10000app
决策账本Decision ledger

先排序,
再花流量。
Rank first,
then allocate traffic.

PMF 先验回答“它天然有多大机会活下来”;行为证据回答“真实用户是否验证了假设”。两者分开,避免把刚上线的零数据误判为失败。The PMF prior asks how likely the product is to survive structurally; behavioral evidence asks whether real users validate the hypothesis. Keeping them separate prevents fresh, zero-data launches from being mislabeled as failures.

评分模型Scoring model

100 分不是 PMF 证明A score of 100 is not proof of PMF

20 + 15 + 15痛点、买家、使用频率pain, buyer clarity, usage frequency
20 + 15经济价值、分发适配economic value, distribution fit
10 + 5差异化、维护杠杆differentiation, maintenance leverage

字段顺序:痛点 / 买家 / 频率 / 价值 / 分发 / 差异化 / 维护。≥70 优先投入;55–69 限时测试;45–54 弱假设;<45 结构性停止。付款或跨月留存可推翻先验。Field order: pain / buyer / frequency / value / distribution / differentiation / maintenance. Prioritize ≥70; time-box 55–69; treat 45–54 as weak; structurally stop below 45. Payment or multi-month retention can overturn the prior.

32 个新候选 · 构建 16 个32 new candidates · 16 built

切线发生在编码前The cut happens before coding

本轮每上线 1 款先生成 2 个候选;按同一七因素模型排序后,只构建前半。CUT 不是上线数小时零收入,而是编码前的结构性选择。Each launch starts with two candidates. They are ranked under one seven-factor model and only the stronger half is built. CUT means a structural choice before coding, not a zero-revenue verdict after a few hours online.

#候选CandidateScore收入路径Revenue path决定Decision依据Basis
17AgentReceipt77会员订阅Membership构建Build持续审计工作流、买家和复核频率最匹配。Advanced to pre-build validation under the same seven-factor PMF prior.
18AIClause75按任务价值匹配qualified lead构建Build高价值合同复核可按合格线索变现。Advanced to pre-build validation under the same seven-factor PMF prior.
19PromptInjectMap74赞助Sponsorship构建BuildAgent 安全有当前分发证据,适合定向赞助。Advanced to pre-build validation under the same seven-factor PMF prior.
20HumanGate73一次性付费One-time payment构建Build高影响批准矩阵有明确团队买家。Advanced to pre-build validation under the same seven-factor PMF prior.
21TraceGap72预充值 API 用量Prepaid API usage构建Build标准化字段让批量 CI 检查可产品化。Advanced to pre-build validation under the same seven-factor PMF prior.
22ShadowAI71按任务价值匹配qualified lead构建Build企业治理痛点高,服务线索价值明确。Advanced to pre-build validation under the same seven-factor PMF prior.
23ModelLicense70一次性付费One-time payment构建Build商业使用前证据请求有高意图。Advanced to pre-build validation under the same seven-factor PMF prior.
24IncidentBrief69一次性付费One-time payment构建Build事件发生时价值高,适合单次响应包。Advanced to pre-build validation under the same seven-factor PMF prior.
25EvidenceClock67会员订阅Membership构建Build证据老化是持续问题,会员逻辑成立待验证。Advanced to pre-build validation under the same seven-factor PMF prior.
26ToolAllowlist66预充值 API 用量Prepaid API usage构建Build权限注册表可反复自动检查。Advanced to pre-build validation under the same seven-factor PMF prior.
27RedTeamPlan65按任务价值匹配qualified lead构建Build高客单安全服务可用初筛获客。Advanced to pre-build validation under the same seven-factor PMF prior.
28EvalDrift64会员订阅Membership构建Build持续评测有频率,但免费替代强。Advanced to pre-build validation under the same seven-factor PMF prior.
29VendorExit63按任务价值匹配qualified lead构建Build迁移高价值但低频,只以线索测试。Advanced to pre-build validation under the same seven-factor PMF prior.
30CostSLO61会员订阅Membership构建Build成本监控高频但产品差异弱。Advanced to pre-build validation under the same seven-factor PMF prior.
31ConsentFlow60一次性付费One-time payment构建Build法规时点可分发,但模板拥挤。Advanced to pre-build validation under the same seven-factor PMF prior.
32LocalOrAPI59联盟分成Affiliate commission构建Build高搜索面但买家和佣金路径弱。Advanced to pre-build validation under the same seven-factor PMF prior.
33BrowserAgentSafety58赞助Sponsorship淘汰Cut问题真实,但与 PromptInjectMap、HumanGate 和 MCPPermit 高度重叠。Cut before coding under the same seven-factor PMF prior.
34VoiceConsent57一次性付费One-time payment淘汰Cut法规需求存在,但语音平台分发和维护成本高。Cut before coding under the same seven-factor PMF prior.
35SyntheticRisk56合格线索Qualified leads淘汰Cut范围过宽,容易变成不可靠的合规判断。Cut before coding under the same seven-factor PMF prior.
36ModelCardGap55一次性付费One-time payment淘汰Cut与 ModelLicense 重叠且免费模板丰富。Cut before coding under the same seven-factor PMF prior.
37AgentMemoryTTL54预充值 API 用量Prepaid API usage淘汰Cut安全真实但单一 TTL 不是充分控制面。Cut before coding under the same seven-factor PMF prior.
38FineTuneOrRAG53联盟分成Affiliate commission淘汰Cut内容竞争拥挤且答案强依赖数据集。Cut before coding under the same seven-factor PMF prior.
39DatasetCardGap52一次性付费One-time payment淘汰Cut专业买家窄、免费文档替代强。Cut before coding under the same seven-factor PMF prior.
40VectorDrift51会员订阅Membership淘汰Cut监测需求存在但需接入真实数据,静态工具价值弱。Cut before coding under the same seven-factor PMF prior.
41ModelBenchmarkPicker50联盟分成Affiliate commission淘汰Cut排行榜拥挤且容易快速过时。Cut before coding under the same seven-factor PMF prior.
42AITranslationQA49预充值 API 用量Prepaid API usage淘汰Cut语言对和领域碎片化,通用工具差异弱。Cut before coding under the same seven-factor PMF prior.
43GPUOrAPI48联盟分成Affiliate commission淘汰Cut与 LocalOrAPI 重叠且实时价格维护重。Cut before coding under the same seven-factor PMF prior.
44ChatbotHandoff47一次性付费One-time payment淘汰Cut平台已有成熟流程,独立模板买家弱。Cut before coding under the same seven-factor PMF prior.
45AIROIStory46一次性付费One-time payment淘汰Cut故事生成低频且容易成为无差异模板。Cut before coding under the same seven-factor PMF prior.
46LLMAccessibility44合格线索Qualified leads淘汰Cut需求重要但范围过大,当前无法提供可信自动判断。Cut before coding under the same seven-factor PMF prior.
47MCPRegistryRFP42按任务价值匹配listing淘汰Cut买家规模和付费路径尚不清晰。Cut before coding under the same seven-factor PMF prior.
48PromptPackShelf34按任务价值匹配ads淘汰Cut可复制、低意图、无持续买家。Cut before coding under the same seven-factor PMF prior.
组合排名Ranked portfolio

逐 App PMF 评分PMF score for every app

App先验分Prior七项分数Seven factors收入 / 证据阶段Revenue / evidence stage评分依据Scoring basis决定Decision
#0034 输入一次方法、状态码、幂等键和副作用描述API Retry Safety
在线Live
88/100
中低置信度medium-low confidence
19 / 14 / 13 / 19 / 12 / 7 / 4 按任务价值匹配team subscription
尚未形成成熟行为证据launched no distribution
RFC 9110 定义 HTTP 方法的安全与幂等语义;AWS SDK 按瞬态、限流、非重试错误、尝试上限、退避和 token bucket 自动决策;Stripe 通过服务端幂等键让写请求安全重放。能力成熟但碎片化,尚不能直接审计任意 API 的业务副作用。PMF prior 88/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.事故成本、购买者和持续频率都明确;规范与 SDK 提供可引用基础,却没有跨供应商审计业务副作用的统一只读层。重构为团队策略订阅并坚持证据不足即人工门禁后,具备可验证 PMF 假设。Retained for a time-boxed test; PMF has not been claimed.
#0130 券商导入归一化器Import Bridge
在线Live
86/100
高置信度high confidence
18 / 14 / 14 / 13 / 13 / 8 / 6 按任务价值匹配one time broker import pack
尚未形成成熟行为证据multi-source hard gates
PMF 先验 86/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 multi-source hard gates;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Ghostfolio and Rotki histories independently show recurring import incompatibility and manual repair. Paid import products prove budget, while local reversible preflight avoids custody and advice.86/100 预判:跨来源导入失败、高成本人工清洗和现有付费导入市场同时成立;不碰账户、成本基础、税务和投资结论。$12 一次处理完整文件。Retained for a time-boxed test; PMF has not been claimed.
#0033 建立一个客户的服务/责任/访问矩阵Desk MSP Boundary
在线Live
88/100
中低置信度medium-low confidence
19 / 15 / 12 / 19 / 12 / 7 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
NIST CSF 2.0 明确建议为供应商、客户和业务伙伴建立责任矩阵,并把共享网络安全责任写入政策和第三方协议;CISA 对 MSP 客户进一步要求在签约前定义 MSP 预期权限和访问级别。Microsoft GDAP 已实现按客户、按角色、统一到期的委派关系,并推荐最小权限与 JIT。现有替代通常是 MSA/SOW、Excel、PSA 文档和各平台权限控制,缺少把“合同责任、操作负责人、实际访问、到期和复核证据”连在一起的轻量跨客户矩阵。PMF prior 88/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.这是由 NIST、CISA 和 Microsoft 控制实践共同支持的高影响问题,买方清楚,争议和过度访问均有直接经济与安全成本。与普通 RACI 不同,把约定责任、实际技术权限、到期和双方复核连成版本化闭环具备差异化;仍需防范法律期待和低频使用,但 MSP 多客户模型足以形成订阅先验,因此高分保留。Retained for a time-boxed test; PMF has not been claimed.
#0110 在浏览器本地检查最多 3 条脱敏 Action/runner 记录的 ownership、runtime、version、hooks、OS/architecture、forced Node24 test、result、fallback、evidence、owner 与 decisionGitHub Actions Node24 Migration Map
在线Live
88/100
中置信度medium confidence
19 / 14 / 15 / 19 / 13 / 5 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
GitHub 官方公告在 2026-05 更新:6 月 16 日起 runners 默认 Node24,Node20 临时 opt-out 只到 2026 秋季移除;Node24 不兼容 macOS 13.4 及更低且不官方支持 ARM32 self-hosted runner。官方 metadata syntax 同时明确 JavaScript Actions 的 node20/node24 runs.using 合约。 Evidence: https://github.blog/changelog/2025-09-19-deprecation-of-node-20-on-github-actions-runners/ · https://docs.github.com/en/actions/reference/workflows-and-actions/metadata-syntaxPMF prior 88/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 88/100。默认切换已发生、最终移除仍有短期紧迫性,Action 维护者/用户/runner 负责人及兼容后果明确;但 runner warning、依赖升级机器人和普通迁移项目是强替代,且时效窗口会衰减,只验证组合迁移证据的付费意愿。Retained for a time-boxed test; PMF has not been claimed.
#0112 GitHub Actions pwn-request checkout 边界交接图GitHub Pwn Request Checkout Boundary
在线Live
92/100
高置信度high confidence
20 / 15 / 14 / 20 / 13 / 7 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 92/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。GitHub is enforcing safer checkout defaults for common fork PR head/merge fetches in privileged pull_request_target and workflow_run paths. Floating supported major tags receive the protection, while fixed pins require updates; git/gh fetches, issue_comment, unrelated third-party sources and explicit unsafe opt-outs remain outside or bypass the narrow protection. GitHub Security Lab independently documents the pwn-request source-to-execution-to-privilege attack path and split-workflow substitute. Evidence: https://github.blog/changelog/2026-06-18-safer-pull_request_target-defaults-for-github-actions-checkout/ · https://securitylab.github.com/resources/github-actions-preventing-pwn-requests/ · https://docs.github.com/en/actions/security-for-github-actions/security-guides/security-hardening-for-github-actions保留为限时验证;尚未声称已获得 PMF。PMF prior 92/100: the migration is current, security pain and accountable platform/repository buyers are explicit, and a cross-path receipt can bridge checkout version, alternative fetch, execution and privilege owners. The wedge is narrower than scanners and GitHub native controls, so it earns a time-boxed launch rather than assumed PMF; no behavior verdict before the preset age and qualified-visit sample.
#0113 GitHub 自托管 Runner 集群新鲜度交接图GitHub Runner Fleet Freshness Map
在线Live
93/100
高置信度high confidence
20 / 15 / 15 / 20 / 13 / 7 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 93/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。GitHub currently sets full enforcement for Data Residency on 2026-07-31 and Enterprise Cloud on 2026-09-25. Version 2.329.0 is only the registration floor; execution eligibility moves with every runner release and a 30-day freshness rule, while critical security updates can pause job queuing immediately. GitHub also states registration audit events are not a complete connected-runner inventory and asks teams to update installation scripts, VM/container images and cached templates. Evidence: https://github.blog/changelog/2026-06-12-github-actions-minimum-version-enforcement-timeline-for-self-hosted-runners/ · https://docs.github.com/en/actions/hosting-your-own-runners/managing-self-hosted-runners/communicating-with-self-hosted-runners · https://docs.github.com/en/rest/actions/self-hosted-runners保留为限时验证;尚未声称已获得 PMF。PMF prior 93/100: the deadline and ongoing moving minimum create direct CI continuity risk for identifiable enterprise platform owners. Native annotations/APIs, ARC, CMDB and image automation are strong substitutes, but the cross-system inventory-update-network-image-workload handoff remains a credible paid wedge. This is only a pre-judgment; behavior cannot be inferred before the age and qualified-visit gates.
#0111 在浏览器本地检查最多 3 条脱敏 cache 路径的 trigger trust、scope、restore/save、external input、trusted consumer、warning、replacement、test、evidence、owner 与 decisionGitHub Untrusted Cache Boundary Map
在线Live
90/100
中置信度medium confidence
20 / 14 / 14 / 20 / 13 / 6 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
GitHub 2026-06-26 官方公告确认:无写权限外部人员可触发且使用共享默认分支 SHA/cache scope 时只发只读 cache token,以阻断 untrusted writer 到 push/schedule 等 trusted restorer 的 cache-poisoning/秘密外泄路径。被限制 save 仅告警且 job 继续、restore 不变,因此还会造成静默性能回归。 Evidence: https://github.blog/changelog/2026-06-26-read-only-actions-cache-for-untrusted-triggers/ · https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflowsPMF prior 90/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 90/100。供应链安全后果高、权限变化新、受影响工作流会在 save 失败后继续,安全迁移与回归排查触发都很强;但 GitHub warning、安全扫描和人工 workflow review 是强替代,只验证跨 workflow trust 证据包的付费意愿。Retained for a time-boxed test; PMF has not been claimed.
#0008 模型退役准备度Sunset Watch
在线Live
78/100
中置信度medium confidence
17 / 12 / 13 / 17 / 12 / 4 / 3 会员订阅Membership
尚未形成成熟行为证据launched no distribution
退役会中断生产且持续变化;买家为使用多模型的工程团队。已有日期表,差异必须来自提醒与迁移工作流。PMF prior 78/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.最高优先:供应商变化持续发生,适合会员情报。Retained for a time-boxed test; PMF has not been claimed.
#0016 Agent 行动收据规划器Agent Receipt
在线Live
77/100
中置信度medium confidence
18 / 14 / 13 / 15 / 10 / 4 / 3 会员订阅Membership
尚未形成成熟行为证据official-guidance + social-demand
NIST/OTel 强调可追溯与持续复核;X 上 Agent 安全内容有真实分发,但尚无本产品付款。PMF prior 77/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance + social-demand. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.高优先:明确的治理/工程工作流和持续复核频率;先验证会员留存。Retained for a time-boxed test; PMF has not been claimed.
#0129 上下文接力Context Relay
在线Live
84/100
高置信度high confidence
18 / 14 / 13 / 12 / 13 / 8 / 6 按任务价值匹配one time local recovery workspace
尚未形成成熟行为证据multi-source hard gates
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 multi-source hard gates;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Claude Code and Codex reports show repeated loss at compaction and session boundaries. The wedge is a human-verified receipt; native client equivalence is the central displacement risk.84/100 预判:重复痛点和可触达开发者成立;原生客户端吸收功能是主要风险。免费生成与复制,$9 一次买断本地历史和 Markdown 导出。Retained for a time-boxed test; PMF has not been claimed.
#0115 处理最多 3 条虚构或脱敏 authority pathnpm Publish Authority Migration Map
在线Live
92/100
高置信度high confidence
20 / 15 / 14 / 20 / 12 / 8 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
npm 的当前一手公告给出两个不同的迁移窗口:2026 年 8 月起,绕过 2FA 的 granular access token 不再能绕过敏感账户、包和组织管理;约 2027 年 1 月起,它也不能直接发布,只保留私有读取和 stage 能力。官方同时提供 trusted publishing OIDC 与 staged publishing 加人工 2FA 批准路径,因此买方痛点不是日期提醒,而是逐包、逐流水线拆分现有权限、目标权限、token 退休、负向测试和恢复责任。 Evidence: https://github.blog/changelog/2026-07-08-npm-install-time-security-and-gat-bypass2fa-deprecation/ · https://github.blog/changelog/2026-05-22-staged-publishing-and-new-install-time-controls-for-npm/PMF prior 92/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 92/100。强制去权窗口、供应链风险和发布中断后果都是真实且当前的,npm maintainer、平台与安全 owner 清晰;但 npm 原生 trusted publishing、staged publishing 和成熟团队内部 runbook 是强替代,所以产品只做跨包迁移证据交接,不宣称执行或安全保证。一次性 18 美元与短期迁移决策匹配,若用户只需要官方说明则没有付费价值。Retained for a time-boxed test; PMF has not been claimed.
#0131 USB 路径诊断器USB Path
在线Live
82/100
中置信度medium confidence
18 / 14 / 11 / 13 / 14 / 7 / 5 按任务价值匹配one time repair handoff pack
尚未形成成熟行为证据multi-source hard gates
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 multi-source hard gates;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。XDA and Android Stack Exchange show repeated mode-specific detection failures. Search reach is strong; willingness to pay is capped by free answers and device fragmentation.82/100 预判:问题重复且长尾搜索明确,但免费论坛替代强。完整诊断永久免费,仅以 $5 维修交接导出测试付费,不执行解锁、刷机或驱动安装。Retained for a time-boxed test; PMF has not been claimed.
#0032 的开工前 briefing 覆盖检查Field Safety Brief
在线Live
77/100
中低置信度medium-low confidence
18 / 13 / 14 / 15 / 10 / 3 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 77/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。OSHA job briefing subjects and multiple commercial field-safety products confirm the workflow. Jobber already supplies required forms, photos, signatures and offline completion, so differentiation and value were reduced from the catalog prior.原始 88 分在竞品复核后下调:完整现场服务套件替代强;只测试无账户窄工具与 $5 可复用模板包。Retained for a time-boxed test; PMF has not been claimed.
#0109 在浏览器本地检查最多 3 条脱敏 policy 记录的 provenance、trust root、image include/exempt、namespace、enforcement、allow/deny、rollback、evidence、owner 与 decisionGitHub Attestation Admission Map
在线Live
87/100
中置信度medium confidence
19 / 14 / 14 / 19 / 12 / 6 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
GitHub 官方文档明确指出仅生成 artifact attestation 不产生安全收益,必须验证;Kubernetes 强制要求 pushed provenance、Sigstore Policy Controller、GitHub TrustRoot、ClusterImagePolicy 与 namespace opt-in。未匹配 include 或同时匹配 include/exempt 会被拒绝,误配有可验证的生产后果。 Evidence: https://docs.github.com/en/actions/how-tos/secure-your-work/use-artifact-attestations/enforce-artifact-attestations · https://docs.github.com/en/actions/concepts/security/artifact-attestationsPMF prior 87/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 87/100。买方、部署触发与阻断/绕过后果清晰;但真实执行已由 GitHub、Sigstore、GitOps、Kyverno/Gatekeeper 等强替代占据,只验证跨系统 rollout 证据包的付费意愿。刚上线与零流量不判通过或失败。Retained for a time-boxed test; PMF has not been claimed.
#0114 npm v12 安装迁移边界交接图npm V12 Install Boundary Map
在线Live
91/100
高置信度high confidence
19 / 14 / 15 / 19 / 13 / 8 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 91/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。npm v12 is now latest and changes install-time defaults: dependency lifecycle scripts and implicit node-gyp no longer run unless approved, git/file/link prepare is covered, and git or remote dependencies default to not being resolved. npm provides approve/deny, project allowScripts, distinct global/npx allow-scripts, strict behavior and dangerous overrides, creating a current migration with both build-break and supply-chain execution consequences. Evidence: https://github.blog/changelog/2026-07-08-npm-install-time-security-and-gat-bypass2fa-deprecation/ · https://github.blog/changelog/2026-06-09-upcoming-breaking-changes-for-npm-v12/ · https://docs.npmjs.com/cli/install/保留为限时验证;尚未声称已获得 PMF。PMF prior 91/100: a current major-version default change affects repeated installs and CI for clear Node/platform/security buyers. Native npm commands, SCA and CI are strong substitutes, but cross-context script/native/source/policy/result evidence is a credible narrow release handoff. Launch is time-boxed and no behavioral verdict is allowed before age plus qualified traffic.
#0036 按组织规模和技术环境生成事实清单Secure Insurance Prep
在线Live
87/100
中低置信度medium-low confidence
18 / 15 / 10 / 20 / 13 / 7 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
保险公司和经纪平台已经能快速报价,Coalition 还结合公开攻击面扫描、持续监控和投保后工具;Beazley 等公开申请表显示 MFA、EDR、备份、培训、邮件安全和数据类型都是具体问题。空档不在替代承保,而在投保企业内部:财务/经纪拿到问题后,要向 MSP、IT、HR 和供应商收集可核验事实,避免猜答、漏答和把“部分启用”写成“全部启用”。PMF prior 87/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.这是日期明确、买方清楚、经济后果直接且渠道就在交易旁边的任务;承保平台解决报价,不解决企业内部跨角色收集可复核事实。只要坚持范围限定和人工签字,15 美元一次性准备包具有较强 PMF 先验。Retained for a time-boxed test; PMF has not been claimed.
#0035 生成并打印付款信息变更政策Secure Payment Fraud
在线Live
87/100
中低置信度medium-low confidence
19 / 14 / 12 / 20 / 12 / 6 / 4 按任务价值匹配usage based
尚未形成成熟行为证据launched no distribution
替代方案从纸面回拨清单、会计师或银行人工核验,到 Zip、Nium 等银行账户验证与 AP 自动化平台。最贴近的 CallbackProof 已把“使用历史可信号码回拨、第二人批准、不可篡改日志和审计 PDF”做成每月 99 美元起的多客户产品,说明工作流与证据可收费,也说明本产品不能只复制清单,必须面向更小团队提供按次、低数据暴露的入口。PMF prior 87/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.这是高损失、触发明确、可验证完成且已有高价竞争产品证明的流程型需求;按次收费能覆盖被月订阅忽略的微型企业。差异化不是检测欺诈,而是以最小敏感数据强制独立回拨和证据闭环,因此保留,但必须靠真实案件复用而非安全内容流量证明 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0012 EU AI Act 范围初筛与人工升级入口Act Intake
在线Live
76/100
中置信度medium confidence
18 / 14 / 9 / 18 / 9 / 4 / 4 按任务价值匹配qualified lead
尚未形成成熟行为证据launched no distribution
法规时点、买家和专业服务客单明确;免费初筛拥挤,价值在合格升级和服务匹配。PMF prior 76/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.高优先:法规时点、买家和高价值服务路径清晰。Retained for a time-boxed test; PMF has not been claimed.
#0118 处理最多 3 条虚构或脱敏 workflow pathGitHub Models Retirement Migration Map
在线Live
91/100
高置信度high confidence
19 / 14 / 14 / 19 / 13 / 8 / 4 按任务价值匹配one time service retirement pack
尚未形成成熟行为证据launched no distribution
GitHub 当前一手公告确认,GitHub Models 将于 2026-07-30 对所有客户完全关闭,playground、model catalog、inference API 和 BYOK 均不再可用;7 月 16 与 23 的 brownout 已让请求暂时返回错误。官方指向 Microsoft Foundry,GitHub 内 AI workflow 则可评估 Copilot,但这两类替代并不自动保留 auth、endpoint、model、request/response、tool/streaming、region/retention、rate/cost 或产品行为。真实任务是逐 workflow 建立可逆的停服迁移证据。 Evidence: https://github.blog/changelog/2026-07-01-github-models-is-being-fully-retired-on-july-30-2026/ · https://docs.github.com/en/github-modelsPMF prior 91/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 91/100。全量硬停服只剩三天,现有活跃 API/BYOK 客户会直接中断,AI/平台/安全/产品/FinOps 买方和经济后果清晰;但需求会在停服后快速衰减,Foundry、Copilot、直接 provider、gateway/SDK 和内部 runbook 都是强替代。采用一次性 20 美元,不做订阅;若搜索和外部完成在停服后消失,立即按预设门槛停止。Retained for a time-boxed test; PMF has not been claimed.
#0017 AI 供应商合同缺口初筛AI Clause
在线Live
75/100
中置信度medium confidence
19 / 14 / 9 / 18 / 8 / 4 / 3 按任务价值匹配qualified lead
尚未形成成熟行为证据official-guidance
合同缺口经济价值高、买家明确;法律边界和合作方质量限制自动化。PMF prior 75/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.高价值买家明确,但收入依赖可信专业合作方与线索质量。Retained for a time-boxed test; PMF has not been claimed.
#0037 生成离职范围清单Desk Leaver Proof
在线Live
86/100
中低置信度medium-low confidence
19 / 14 / 11 / 20 / 11 / 7 / 4 按任务价值匹配usage based
尚未形成成熟行为证据launched no distribution
Microsoft Entra Lifecycle Workflows 能禁用账号、移除群组/Teams/许可证、结束访问包并保留工作流历史,但主要覆盖 Microsoft 身份边界。1Password SaaS Manager 等更完整产品把 SaaS 账号、设备锁定/擦除、归还任务和审计证据连接,Offboarder 也专门销售跨混合环境的撤权证明。竞争成熟,但小团队仍可能缺少跨 SaaS、共享链接与实物回收的统一完成证据。PMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.离职撤权是官方明确的高风险动作,且成熟产品已把审计证据作为卖点。与通用清单不同,本候选可在不执行高风险操作的前提下汇总跨身份、SaaS、共享访问和设备的可复核完成证明,按事件收费与小团队匹配,因此保留;但必须用真实遗漏发现和专业抽查验证。Retained for a time-boxed test; PMF has not been claimed.
#0038 导入最多三个供应商或一百个席位的购买、分配、活跃 CSVDesk License Recon
在线Live
86/100
中低置信度medium-low confidence
18 / 14 / 13 / 20 / 11 / 6 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
Microsoft 365 与 Google Workspace 管理后台已经分别提供许可证分配、用户活跃和可导出报表;1Password SaaS Manager、Productiv、Zylo、BetterCloud 等完整 SaaS 管理平台还能连接应用、发现使用情况、自动回收席位并管理续约。竞品验证了需求与预算,但多为面向中大型组织、需要集成和询价的综合平台,轻量 CSV 对账工具仍可服务尚未准备采购完整平台的小型 IT 团队与 MSP。PMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.这是买方、频次和经济价值都较清楚的成本控制任务,官方管理后台与成熟 SaaS 管理平台共同证明行为存在。小团队可通过低权限、本地 CSV、跨供应商对账获得足够差异化,但数据口径不一致与企业竞品强势会压低分发和差异化,因此保留验证而非视为已获得 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0108 在浏览器本地检查最多 3 条脱敏 trust 记录的仓库生命周期、subject 格式、audience、environment/ref、云条件、最小权限、非生产测试、回退、证据、负责人和决定GitHub OIDC Trust Migration Map
在线Live
86/100
中置信度medium confidence
19 / 14 / 14 / 19 / 12 / 5 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
GitHub 官方文档确认 OIDC 以 aud/sub 等 claims 约束云信任;2026-07-15 后新仓库默认使用包含 owner/repository ID 的 immutable subject,之后的仓库重命名或转移也进入新格式,形成明确且高后果的迁移触发。原生 GitHub 设置、各云 IAM、IaC 与安全平台仍是强替代。 Evidence: https://docs.github.com/en/actions/reference/security/oidc · https://docs.github.com/en/actions/concepts/security/openid-connectPMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 86/100。买方、事件触发和部署/权限后果明确,但产品不解析真实 token 或生成策略;仅验证跨 GitHub 与多云系统之间的无秘密证据交接是否值得单独付费。刚上线及零流量不作为通过或失败证据。Retained for a time-boxed test; PMF has not been claimed.
#0018 Prompt injection 暴露面地图Prompt Inject Map
在线Live
74/100
中置信度medium confidence
18 / 13 / 11 / 17 / 10 / 3 / 2 赞助Sponsorship
尚未形成成熟行为证据official-guidance + social-demand
OWASP 2026 Agentic 风险与 X 安全内容互动证明分发面;竞品拥挤。PMF prior 74/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance + social-demand. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.安全内容在 X / GitHub 有分发证据;赞助需要先证明定向流量。Retained for a time-boxed test; PMF has not been claimed.
#0125 在浏览器本地映射最多 3 条脱敏 exposure 路径的 attribution、public surface、secret class/validity、external ownership、service owner、revoke/rotate、content removal、notification、audit、prevention、recovery、证据和 ownerGitHub Public Secret Exposure Response Map
在线Live
88/100
中置信度medium confidence
19 / 14 / 12 / 20 / 13 / 7 / 3 按任务价值匹配one time public exposure response pack
尚未形成成熟行为证据launched no distribution
PMF 先验 88/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。PMF prior 88/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.88/100:公开凭据可能被即时利用,平台归因首次形成明确企业响应队列,AppSec/IAM 买家和经济后果强;但原生 alert/recommendations、provider revocation、SIEM/SOAR 与 incident runbook 极强,因此只验证 external ownership 到 service recovery 的跨边界证据,先排队不占主动流量席位。Retained for a time-boxed test; PMF has not been claimed.
#0013 MCP 权限面检查器MCP Permit
在线Live
73/100
中置信度medium confidence
17 / 13 / 11 / 16 / 10 / 3 / 3 赞助Sponsorship
尚未形成成熟行为证据launched no distribution
MCP 权限风险真实且安全买家清晰;免费扫描器多,先用可解释检查和赞助验证。PMF prior 73/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.高优先:安全买家明确;扫描器拥挤,先测试教育流量和赞助。Retained for a time-boxed test; PMF has not been claimed.
#0046 供应商接入分级工具Secure Vendor Risk
在线Live
85/100
中低置信度medium-low confidence
18 / 14 / 12 / 19 / 11 / 7 / 4 按任务价值匹配usage based
尚未形成成熟行为证据launched no distribution
PMF 先验 85/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。NIST treats supplier due diligence as a real acquisition decision process and provides C-SCRM guidance; CISA publishes a free SMB operational template. Vanta and UpGuard prove a mature paid category spanning questionnaires, evidence, findings, remediation and recurring monitoring, but their full-platform scope and public UpGuard entry price leave a plausible low-end wedge for a small team assessing one vendor occasionally. The counterevidence is strong: free templates and existing compliance platforms can fully substitute for many buyers, and a browser form cannot certify controls. Therefore the product is deliberately limited to declared relationship tiering and an editable evidence request, with no external score, scan, automated approval or legal conclusion. Evidence: https://csrc.nist.gov/pubs/sp/1326/final · https://csrc.nist.gov/pubs/sp/1305/final · https://www.cisa.gov/sites/default/files/2025-08/Operationalizing_the_Vendor_SCRM_Template_for_SMBs_2025_Final_508.pdf · https://help.vanta.com/en/articles/11345550-conducting-vendor-security-assessments · https://www.upguard.com/pricing/vendor-risk保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 85/100. Keep as a testable narrow candidate because the acquisition trigger, accountable security/procurement buyer and downside of unmanaged data/access dependencies are clear, while $5 per vendor matches an occasional one-off job. The defensible wedge is not security scoring: it is a local, transparent intake for organizations too small to justify enterprise TPRM. This remains a prior, not PMF; free CISA/NIST material and Vanta/UpGuard are powerful substitutes, so survival requires real evidence requests and non-affiliated purchases after the time/sample gate.
#0083 仅使用别名的跨工具远程访问权限复核Desk Remote Risk
在线Live
84/100
中低置信度medium-low confidence
18 / 14 / 13 / 18 / 12 / 5 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA explicitly treats remote administration and RMM as a high-impact attack surface, calls for MFA on remotely accessible and vendor or maintenance accounts, and recommends auditing remote-access tools and reviewing execution logs. NIST least-privilege and remote-access controls reinforce periodic privilege review. Native vendor consoles and PAM/MSP platforms remain strong substitutes, so the narrow opportunity is a browser-local cross-tool normalization and human evidence queue for small MSP portfolios, not control enforcement. Evidence: https://www.cisa.gov/resources-tools/resources/guide-securing-remote-access-software · https://www.cisa.gov/sites/default/files/publications/CISA_CPG_CHECKLIST_508c.pdf · https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-136a · https://csrc.nist.gov/pubs/sp/800/53/r5/upd1/final保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 84/100: the recurring cross-tool permission review has a clear MSP and IT buyer, material downside and quarterly reuse, while local redacted exports create a safe low-cost wedge. Differentiation is only moderate because native consoles and PAM products own the real controls, and the tool cannot find shadow RMM. Keep live for a 90-day evidence test; zero traffic or early age is not failure.
#0048 人类与 LLM 裁判一致性分析Eval Judge Calibrate
在线Live
84/100
中低置信度medium-low confidence
18 / 14 / 14 / 18 / 12 / 4 / 4 按任务价值匹配prepaid report
尚未形成成熟行为证据launched no distribution
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Research shows the task is real but not solved by one agreement percentage. ACL studies report that human–LLM judge alignment varies by model, evaluated property, language, setting and human expertise; high raw agreement can still hide materially different scoring behavior and biases. LangChain and Arize are strong substitutes because calibration belongs inside mature evaluation/observability workflows. The narrow wedge is a portable, browser-local analysis for teams holding only redacted item IDs and labels, with explicit confusion, prevalence, slices, disagreements and version records. It cannot validate the underlying human labels or the judge outside the sampled task. Evidence: https://aclanthology.org/2025.gem-1.33/ · https://aclanthology.org/2025.acl-short.20/ · https://aclanthology.org/2024.emnlp-main.451/ · https://www.langchain.com/resources/llm-as-a-judge · https://arize.com/blog/measuring-human-llm-judge-alignment/保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 84/100. Judge calibration is frequent, has a clear evaluation/AI-platform buyer and can alter release decisions. A $9 one-report unit matches periodic recalibration better than an assumed subscription. Differentiation is modest and incumbent platforms can add the same metrics, so the candidate survives only if privacy/local portability and disagreement adjudication change a real judge decision and produce repeat calibration behavior plus non-affiliated payments.
#0019 Agent 人工批准矩阵Human Gate
在线Live
73/100
中置信度medium confidence
18 / 13 / 12 / 17 / 8 / 3 / 2 一次性付费One-time payment
尚未形成成熟行为证据official-guidance
高影响 Agent 的人工监督需求清晰;模板复购弱,定价为一次性团队包。PMF prior 73/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.问题清晰且高影响,但模板复购低,匹配一次性团队决策包。Retained for a time-boxed test; PMF has not been claimed.
#0103 Google Play 购买生命周期交接图Mobile Play Billing State Route
在线Live
84/100
高置信度high confidence
18 / 13 / 14 / 19 / 12 / 5 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google currently recommends a backend purchase-status system using real-time developer notifications as lifecycle signals and Play Developer APIs for current purchase truth. Pending, purchased, acknowledgement, consumption, cancellation, refund and void states can change entitlement and revenue, and failure to acknowledge eligible purchases within three days can cause automatic refund and revocation. The risk and buyer are real, but official libraries and subscription platforms are strong substitutes. Evidence: https://developer.android.com/google/play/billing/lifecycle · https://developer.android.com/google/play/billing/lifecycle/one-time · https://developer.android.com/google/play/billing/release-notes保留为限时验证;尚未声称已获得 PMF。PMF prior 84/100: advance only as a narrow one-time cross-endpoint release evidence pack. Payment state, entitlement, acknowledgement or consumption, duplicates and reversals are economically important and recur across Android billing releases. Differentiation is limited because Play Billing, Developer APIs, testing and RevenueCat execute the workflow, so fixed behavior and payment gates control further investment.
#0106 Google Play 目标 API 截止日期交接图Mobile Target API Deadline Map
在线Live
84/100
中低置信度medium-low confidence
18 / 13 / 12 / 19 / 13 / 6 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google current guidance sets August 31, 2026 target API requirements for new apps, updates and existing-app availability, with form-factor differences and a November 1 extension path. Play Console owns authoritative status; the wedge is portfolio-level date, test and owner handoff. Evidence: https://support.google.com/googleplay/android-developer/answer/11926878?hl=en · https://developer.android.com/google/play/requirements/target-sdk保留为限时验证;尚未声称已获得 PMF。The deadline and discoverability consequences create a high-value release trigger across multiple form factors, but Console warnings and ordinary project management are strong substitutes; only the cross-App evidence handoff deserves a small one-time test.
#0047 任务级单位经济工具Token Unit Cost
在线Live
84/100
中低置信度medium-low confidence
19 / 15 / 15 / 19 / 9 / 4 / 3 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据launched no distribution
PMF 先验 84/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。The FinOps Foundation explicitly treats cost per transaction, case resolved and other business outcome as actionable unit economics, and its AI workload guidance names engineering, finance and business stakeholders. Google Cloud likewise describes combining allocated cost with a value metric. Langfuse is strong counterevidence: it already records model usage/cost and exposes grouped metrics, so another per-call token calculator has no wedge. The testable gap is the denominator: joining every retry, model/tool step and failed/unknown outcome to one redacted business task. This only creates value if buyers have a reliable outcome source and the metric changes routing, product, pricing or retry decisions. Evidence: https://www.finops.org/framework/capabilities/unit-economics/ · https://www.finops.org/wg/cost-estimation-of-ai-workloads/ · https://langfuse.com/docs/observability/features/token-and-cost-tracking · https://langfuse.com/docs/metrics/overview · https://cloud.google.com/blog/topics/cost-management/leveraging-cloud-finops-to-measure-the-business-value-of-cloud保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 84/100. The pain, buyer, frequency and economic value are unusually clear, while prepaid processing matches variable event volume. Differentiation is narrow and easy for observability incumbents to absorb; therefore this is not presumed PMF. It survives only as a privacy-minimized, portable task/outcome reconciliation layer and must prove manual-ledger accuracy, reliable success definitions, decision changes and non-affiliated payments after the time/sample gate.
#0004 AI 数据留存检查器Retention Lens
在线Live
72/100
中置信度medium confidence
17 / 13 / 12 / 16 / 7 / 3 / 4 一次性付费One-time payment
分发前Pre-distribution
企业治理有预算,但当前静态包必须向持续监控、资产清单或线索路径演进。PMF prior 72/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: pre_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.优先测试并寻找持续监控转型:企业治理买家清晰。Retained for a time-boxed test; PMF has not been claimed.
#0049 周期扣费候选检测Cash Subscription Recon
在线Live
83/100
中低置信度medium-low confidence
19 / 14 / 13 / 20 / 11 / 3 / 3 按任务价值匹配one time audit
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Ramp, Brex and Rocket Money are strong counterevidence: connected spend platforms and consumer tools already identify and act on recurring expenses with richer context. Stripe also shows that merchant-side systems hold authoritative subscription lifecycle and pricing. The economic pain is nevertheless direct and the buyer is clear; a small organization may have only a CSV across cards/accounts and no spend-management suite. The narrow wedge is local, explainable cadence and price-step candidate detection followed by owner/use/contract confirmation. A transaction pattern alone cannot prove a subscription, duplicate, unused license, price increase, cancellability or savings. Evidence: https://support.ramp.com/ramp-savings-insights/ · https://ramp.com/blog/unused-software-subscriptions · https://www.rocketmoney.com/faq · https://www.brex.com/product/expense-management · https://stripe.com/billing/subscriptions保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 83/100. Direct possible savings, a finance/operations buyer and a periodic audit trigger support a $19 one-time review. Competition is extremely strong and differentiation is only a privacy-preserving CSV wedge, so this is not assumed PMF. It must prove high candidate precision, real owner follow-up, verified actions and non-affiliated payments after the time/sample gate; clicks or annualized candidate totals do not count as value.
#0051 重复工单候选分析Desk Repeat Issue
在线Live
83/100
高置信度high confidence
18 / 14 / 13 / 18 / 11 / 5 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Atlassian and Freshservice independently validate recurring-incident problem management, linked investigation, known-error and follow-up workflows. They are also strong substitutes, so the narrow hypothesis is browser-local cross-export triage for small service desks and MSPs that cannot justify replacing their ITSM; lexical similarity is explicitly not treated as root cause. Evidence: https://support.atlassian.com/jira-service-management-cloud/docs/best-practises-for-problem-management/ · https://support.atlassian.com/jira-software-cloud/docs/manage-incidents-in-jira/ · https://www.freshworks.com/freshservice/it-service-management/problem-management/ · https://www.freshworks.com/itil/problem-management/保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 83/100. Recurring tickets consume measurable service-desk time, a service desk manager can act, and monthly before/after recurrence supports repeat use. However Jira Service Management and Freshservice already cover native problem workflows, so differentiation is only local cross-export explainability and repair-ROI evidence. Retain for a gated test, not as proven PMF.
#0124 在浏览器本地映射最多 3 条脱敏处置路径的 incident scope、account mode、target/action、credential coverage/exclusions、automation dependency、audit evidence、replacement、SSO reauthorization、recovery test、证据和 ownerGitHub Credential Revocation Recovery Map
在线Live
87/100
中置信度medium confidence
19 / 14 / 11 / 20 / 12 / 8 / 3 按任务价值匹配one time credential recovery pack
尚未形成成熟行为证据launched no distribution
GitHub 已提供 enterprise 级按用户、凭据类型或全体的 SSO authorization 撤销,以及 EMU credential 删除;官方文档明确动作不可逆、credential coverage 有排除项、批量操作会破坏自动化,并通过 audit events 与 email 留证。 Evidence: https://github.blog/changelog/2026-06-24-self-service-credential-revocation-for-incident-response/ · https://docs.github.com/en/enterprise-cloud@latest/admin/managing-iam/respond-to-incidents/revoke-authorizations-or-tokens · https://docs.github.com/en/authentication/keeping-your-account-and-data-secure/revoking-your-credentialsPMF prior 87/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.87/100:不可逆处置、企业自动化爆炸半径和明确 IAM/SecOps 买家形成高经济价值先验;但 incident 低频,GitHub 原生 UI/API、SIEM/SOAR、PAM 和 runbook 极强,因此只验证 action coverage、排除项到安全恢复的跨 owner 证据,先排队不占主动流量席位。Retained for a time-boxed test; PMF has not been claimed.
#0050 双项目资本比较Green Project ROI
在线Live
83/100
中低置信度medium-low confidence
18 / 13 / 11 / 20 / 11 / 6 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。ENERGY STAR and ORNL provide strong free substitutes for financial and energy-project ROI calculations, so another simple payback formula has little product value. DOE nevertheless documents a real decision and evidence problem: savings are not directly measurable; behavior, weather, sampling, measurement and simplifying assumptions create uncertainty, and M&V rigor should match performance risk and expected savings. The testable wedge is a small-facility cross-quote memo that forces identical scope, low/base/high inputs, transparent NPV and a post-installation evidence plan. It does not estimate engineering performance or incentive eligibility. Evidence: https://www.energystar.gov/buildings/save-energy-commercial-buildings/finance-projects/economic-and-financial-calculators · https://www.energy.gov/femp/types-energy-and-water-cost-savings-can-be-used-pay-project · https://www.energy.gov/cmei/femp/measurement-and-verification-federal-energy-savings-performance-contracts · https://www.energy.gov/cmei/femp/measurement-and-verification-activities-required-energy-savings-performance-contract · https://industrialresources.ornl.gov/justifi/home/保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 83/100. Capital approval ties the task to large economic consequences and a clear facilities/operations/procurement buyer; $12 once matches a project decision better than an assumed subscription. Free official calculators are major counterevidence. The candidate survives only if the comparable-assumption memo—not the formula—helps real buyers revise or approve quotes and produces non-affiliated payments after 60 days and the sample floor.
#0020 Agent trace 完整度检查器Trace Gap
在线Live
72/100
中置信度medium confidence
17 / 13 / 13 / 16 / 7 / 3 / 3 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据standards
OTel 已定义 GenAI/Agent 字段,CI 检查有重复工作流;需证明 API 而非免费替代。PMF prior 72/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: standards. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.批量 CI/注册表工作流适配 API credits;先验证自动化需求。Retained for a time-boxed test; PMF has not been claimed.
#0085 品牌事实源Discover AI Answer Claim Index
在线Live
83/100
高置信度high confidence
17 / 13 / 13 / 16 / 15 / 6 / 3 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google documents that AI-search eligibility uses normal Search fundamentals, requires indexed snippet-eligible pages, and needs no special AI file or schema. Profound already sells claim extraction, source-of-truth comparison and citation attribution, proving enterprise demand and a strong substitute. The remaining small-team wedge is recurring claim-to-source-to-owner governance before monitoring, not an AI-ranking promise. Evidence: https://developers.google.com/search/docs/appearance/ai-features · https://developers.google.com/search/docs/appearance/structured-data/organization · https://developers.google.com/search/docs/fundamentals/creating-helpful-content · https://help.tryprofound.com/articles/5793584301-about-factcheck · https://geolens.info/保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 83/100: price, policy, availability and capability conflicts recur and can mislead customers before AI discovery is even measured. A $9 membership matches versioning, ownership and expiry work better than credits. Differentiation is only moderate because enterprise GEO suites already cover downstream monitoring; static sites may not return monthly, so second-version behavior is the key falsification test.
#0102 App Store Server Notifications V2 场景交接图Mobile Subscription Event Route
在线Live
83/100
高置信度high confidence
18 / 13 / 13 / 19 / 12 / 5 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Apple currently assigns developers responsibility for interpreting and responding to App Store Server Notifications V2 across purchases, renewals, billing failures, grace periods, refunds and other lifecycle events. Production failures can retry and V2 history can recover attempted deliveries, while current-state APIs remain necessary. The buyer and economic risk are real, but Apple libraries and RevenueCat are strong substitutes; the narrow wedge is cross-team event-to-entitlement evidence. Evidence: https://developer.apple.com/documentation/AppStoreServerNotifications · https://developer.apple.com/documentation/AppStoreServerNotifications/responding-to-app-store-server-notifications · https://developer.apple.com/documentation/appstoreserverapi/get-notification-history · https://www.revenuecat.com/docs/integrations/webhooks/event-flows保留为限时验证;尚未声称已获得 PMF。PMF prior 83/100: advance as a narrow one-time release evidence pack, never as a JWS verifier or subscription backend. Events directly affect customer access and revenue, and duplicate, ordering and outage recovery decisions recur across releases. Differentiation remains modest because official libraries, current-state APIs and subscription platforms cover execution, so behavior and payment must meet fixed gates before more investment.
#0011 AI 数据路径规划器Private Route
在线Live
71/100
中置信度medium confidence
16 / 12 / 10 / 16 / 9 / 4 / 4 联盟分成Affiliate commission
尚未形成成熟行为证据launched no distribution
数据路由影响合同与架构;官方资料分散,透明推荐可验证购买前研究意图。PMF prior 71/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.优先测试:高价值问题,可用透明推荐验证出站意图。Retained for a time-boxed test; PMF has not been claimed.
#0084 仅使用别名的恢复演练规划Secure Recovery Test
在线Live
83/100
中低置信度medium-low confidence
18 / 14 / 8 / 18 / 12 / 8 / 5 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA recommends maintaining offline encrypted backups and regularly testing availability and integrity in disaster-recovery scenarios. NIST contingency guidance includes recovery priorities, procedures, testing, training and exercises; current NIST OT guidance likewise links backup testing to recovery exercises. The pain and buyer are real, but vendor consoles, free official guidance, MSPs and continuity consultants are strong substitutes, so the product is deliberately an alias-only planning/evidence artifact rather than a backup or recovery engine. Evidence: https://www.cisa.gov/stopransomware/ransomware-guide · https://csrc.nist.gov/pubs/sp/800/34/r1/upd1/final · https://csrc.nist.gov/topics/security-and-privacy/security-programs-and-operations/contingency-planning · https://csrc.nist.gov/pubs/sp/1339/final保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 83/100: a failed recovery can create severe downtime and authorized IT/MSP owners have a clear one-time exercise job. The narrow preflight, dependency, validation, stop and rollback artifact is safely deliverable without credentials, but exercise frequency is low and native/vendor services constrain repeat value. Keep live for a 90-day behavioral test; do not treat launch age or zero traffic as failure.
#0105 Google Play 数据安全组件交接图Mobile Play Data Safety Map
在线Live
83/100
中低置信度medium-low confidence
18 / 13 / 13 / 18 / 13 / 5 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 83/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google current guidance makes developers responsible for complete Data safety declarations covering App code and third-party SDK behavior, with updates when practices change. Play Console and SDK Index are strong native surfaces; the testable wedge is build-specific cross-component evidence ownership. Evidence: https://support.google.com/googleplay/android-developer/answer/10787469?hl=en · https://developer.android.com/privacy-and-security/declare-data-use · https://play.google.com/sdks保留为限时验证;尚未声称已获得 PMF。SDK, permission, account and deletion-path changes create recurring declaration risk and fragmented ownership, but native Play tooling and scanners already dominate observation and submission; only a narrow one-time release evidence handoff deserves testing.
#0116 处理最多 3 条虚构或脱敏 recovery pathnpm Read Only Recovery Boundary Map
在线Live
86/100
高置信度high confidence
19 / 14 / 11 / 19 / 12 / 8 / 3 按任务价值匹配one time recovery evidence pack
尚未形成成熟行为证据launched no distribution
npm 当前一手公告确认,高影响账户在更改邮箱或使用 2FA recovery code 后自动进入 72 小时只读;安装、下载、浏览组织/团队和账户/包设置仍可用,但发布、token 管理、package visibility、组织与团队成员变更暂停,期满自动恢复且无需再次确认。真实任务不是倒计时,而是区分允许/暂停动作、在不绕过保护的前提下安排关键发布连续性、意外触发时升级 Support,并在自动恢复后人工复核写权限与旧路径。 Evidence: https://github.blog/changelog/2026-06-25-npm-adds-preventive-account-protection-for-high-impact-accounts/ · https://docs.npmjs.com/configuring-two-factor-authentication/PMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 86/100。供应链后果、发布阻断和高影响维护者买方都很清晰,但事件相对低频,npm 原生保护、旧邮箱 alert、Support、trusted/staged publishing 与成熟 incident/release runbook 是强替代。一次性 15 美元只在跨 package 连续性和恢复复核证据交接中有有限付费缝隙;若用户只要状态和倒计时则没有独立价值。Retained for a time-boxed test; PMF has not been claimed.
#0021 Shadow AI 资产初筛Shadow AI
在线Live
71/100
中置信度medium confidence
18 / 13 / 10 / 17 / 7 / 4 / 2 按任务价值匹配qualified lead
尚未形成成熟行为证据official-guidance
治理痛点强、企业买家清晰,但销售周期和资产发现交付重。PMF prior 71/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.企业痛点高但销售周期长;只在有服务合作方时扩大分发。Retained for a time-boxed test; PMF has not been claimed.
#0052 逾期发票下一步队列Cash Late Invoice
在线Live
82/100
中置信度medium confidence
19 / 15 / 15 / 20 / 8 / 2 / 3 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。QuickBooks and Xero validate the everyday aging and reminder workflow while Chaser and Upflow demonstrate a paid specialist category. The same sources are strong substitutes: most connected accounting users already have authoritative balances and reminder history. The narrow test is a no-login, browser-local, transparent next-action queue for small teams that reject full AR automation. Evidence: https://quickbooks.intuit.com/learn-support/en-us/help-article/accounts-receivable-reports/run-accounts-receivable-aging-report/L4N7PC2hg_US_en_US · https://central.xero.com/s/article/How-invoice-reminders-work · https://www.chaserhq.com/chaser-pricing · https://upflow.io/pricing保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Late payment is frequent, economically material and owned by a clear finance buyer, but differentiation is extremely weak because accounting suites and AR platforms already solve the authoritative workflow. Retain only as a low-friction local CSV wedge; disclosed scoring, dispute/promise holds, recorded next actions and real payment—not clicks—must prove value.
#0053 项目贡献毛利对账Cash Project Margin
在线Live
82/100
中置信度medium confidence
19 / 14 / 13 / 20 / 10 / 3 / 3 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Xero Projects, QuickBooks Projects and Float validate that time cost, direct expense, revenue and margin affect project decisions, while also proving that the formula and connected workflow are mature substitutes. The narrow hypothesis is a no-migration browser-local join for small agencies, consultants and service teams whose invoice, expense and time exports remain fragmented. Evidence: https://central.xero.com/s/article/About-Xero-Projects · https://www.xero.com/us/accounting-software/track-projects/ · https://quickbooks.intuit.com/learn-support/en-us/help-article/manage-projects/compare-project-cost-estimates-vs-actuals-online/L9oxsLKqN_US_en_US · https://support.float.com/en/articles/12586520-understanding-margin-in-float保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Project contribution directly affects pricing, staffing and scope, with a clear owner and monthly reuse, but native accounting/project platforms already dominate and the arithmetic is commodity. Retain only for the cross-CSV missing-labor-cost audit; verified corrections, next-month return and real payment must distinguish it from a spreadsheet.
#0123 在浏览器本地映射最多 3 条脱敏 billing visibility 路径的旧用途、account scope、AI usage source、group/filter/export、budget、cost center、credit pool、native replacement、reconciliation、cutover、证据和 ownerCopilot Billing Preview Retirement Map
在线Live
86/100
中置信度medium confidence
18 / 14 / 12 / 19 / 13 / 7 / 3 按任务价值匹配one time billing visibility migration pack
尚未形成成熟行为证据launched no distribution
PMF 先验 86/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。PMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.86/100:硬退役日期、费用可见性和明确的 enterprise billing/FinOps 买家形成强先验,旧 app 的覆盖缺口由 GitHub 直接确认;但原生 billing settings、reports/API、FinOps/BI 和月结 runbook 极强,因此仅验证跨用途对账与切换证据,先排队不占主动流量席位。Retained for a time-boxed test; PMF has not been claimed.
#0055 销售佣金账单预检Deal Comp Proof
在线Live
82/100
中置信度medium confidence
18 / 14 / 12 / 19 / 10 / 6 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Salesforce Spiff and Everstage validate paid demand for transparent commission calculations, statements, audit trails and dispute workflows, while proving strong employer-purchased enterprise substitutes. The narrow hypothesis is a private local preflight for a representative or small team without ICM access: map one manually entered rule and source evidence before asking an authorized reviewer a factual question. Evidence: https://www.salesforce.com/sales/incentive-compensation-management/pricing/ · https://www.everstage.com/ · https://www.everstage.com/commission-tracker-software · https://help.salesforce.com/s/articleView?id=sales.spiff_statements_rep.htm&type=5保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Commission opacity and errors have direct personal and operational cost, monthly or quarterly triggers and a recognizable buyer, while enterprise platforms leave a low-cost independent preflight gap. Differentiation remains narrow and legal/payroll expectations are dangerous; retain only as evidence/question preparation, never plan interpretation or a wage determination.
#0121 在浏览器本地检查最多 3 条脱敏 closed-alert history query 的 cloud scope、state/age、UI/API dependency、archive role/export、fidelity/region、downstream import、test、cutover、evidence 与 ownerDependabot Alert Archive Transition Map
在线Live
86/100
中置信度medium confidence
18 / 14 / 13 / 19 / 12 / 7 / 3 按任务价值匹配one time security history transition pack
尚未形成成熟行为证据launched no distribution
PMF 先验 86/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。PMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.86/100 PMF 预判:硬日期、审计与 remediation history 静默缺失、清晰 AppSec/GRC 买家形成强先验;但原生 archive CSV 和成熟 SIEM/GRC 替代很强,只验证跨 UI/API、archive 和下游查询的证据连续性。Retained for a time-boxed test; PMF has not been claimed.
#0056 终端关键路径覆盖矩阵Desk Backup User
在线Live
82/100
中置信度medium confidence
18 / 14 / 12 / 19 / 10 / 5 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA validates protected recovery copies and restoration planning, while Microsoft documents the useful but bounded Known Folder and OneDrive redirection scope. Backup agents and sync products are strong native substitutes. The narrow independent hypothesis is an MSP/IT governance matrix that aligns business-critical endpoint paths with declared roots, exclusions, separate protection layers and human-observed sample restores. Evidence: https://www.cisa.gov/stopransomware/ransomware-guide · https://learn.microsoft.com/en-us/windows/win32/shell/working-with-known-folders · https://learn.microsoft.com/en-us/sharepoint/redirect-known-folders · https://learn.microsoft.com/en-us/troubleshoot/sharepoint/lists-and-libraries/determine-known-folder-move-eligibility保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Missing critical user files can stop work and the IT/MSP buyer can review scope monthly and restore evidence quarterly, but native backup/sync products own authoritative agent state. Retain only for cross-product path-governance evidence; prefix matching never proves backup or recovery and the product remains distinct from CloudBackupReach system-level failure-domain/RPO/RTO review.
#0119 在浏览器本地检查最多 3 条脱敏 GHES instance 路径的版本线、最低补丁状态、拓扑、bundle/upload route、维护窗口、生成负载、网络、Support fallback、演练、证据与 ownerGHES Support Bundle Upload Readiness Map
在线Live
86/100
中置信度medium confidence
18 / 14 / 12 / 19 / 12 / 7 / 4 按任务价值匹配one time support continuity pack
尚未形成成熟行为证据launched no distribution
GitHub 当前公告确认 2026-08-18 起拒绝未达到五条最低补丁线的旧 GHES appliance 的三类命令行 support bundle 上传;官方文档同时证明 bundle 生成、拓扑、资源、隐私、网络与 Support 路径需要真实管理员判断。原生文档、release notes、CMDB 和 Support 是强替代,窄缝仅是跨实例支持连续性交接。 Evidence: https://github.blog/changelog/2026-07-22-upcoming-ghes-change-impacting-uploading-support-bundles/ · https://docs.github.com/en/enterprise-server@3.17/admin/monitoring-and-managing-your-instance/monitoring-your-instance/about-support-bundles · https://docs.github.com/en/enterprise-server@3.20/support/contacting-github-support/providing-data-to-github-supportPMF prior 86/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.86/100 PMF 预判:明确拒绝日期、企业事故恢复后果、五条补丁线与清晰 GHES 管理员买家构成强触发;但客群窄、任务低频且原生文档、CMDB 和 GitHub Support 替代强,因此只验证一次性跨实例 readiness 证据包,不把合规或平台要求误当付费需求。Retained for a time-boxed test; PMF has not been claimed.
#0057 多地点公用事业账单周期对账Green Utility Recon
在线Live
82/100
中置信度medium confidence
18 / 13 / 13 / 18 / 11 / 5 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。ENERGY STAR Portfolio Manager provides a strong free structured property/meter substitute, while EnergyCAP validates enterprise willingness to pay for bill capture, missing/duplicate audits and accounting workflow. The narrow hypothesis is a low-cost CSV-only exact-date interval checker for 5–30-site operators who already have exports but lack a full utility platform. Evidence: https://portfoliomanager.energystar.gov/pm/dataCollectionWorksheet · https://www.energycap.com/utility-bill-energy-management-software/ · https://helpcenter.energycap.com/um/bill_capture_um/audits保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Billing-period gaps can block recurring property/facility month close and the owner and value are clear, but free Portfolio Manager and enterprise utility platforms strongly constrain differentiation. Retain only for transparent [start,end) gap/overlap/duplicate reconciliation; it does not identify tariff or billing errors, connect accounts or promise savings.
#0022 开放模型许可证证据初筛Model License
在线Live
70/100
中置信度medium confidence
17 / 13 / 8 / 18 / 8 / 4 / 2 一次性付费One-time payment
尚未形成成熟行为证据official-documentation
模型卡与许可证证据是商业使用前置;低频且必须避免法律判断。PMF prior 70/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-documentation. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.商业使用决策价值高;避免法律结论,出售证据请求包。Retained for a time-boxed test; PMF has not been claimed.
#0054 面向小企业的事前事件响应行动卡Secure Incident Plan
在线Live
82/100
中置信度medium confidence
18 / 14 / 9 / 20 / 11 / 6 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA and FTC validate that small businesses need assigned response roles, current external contacts, evidence/action records, communications, continuity and practiced plans. Their free resources are strong substitutes for generic templates. The narrow paid hypothesis is an approved offline action-card and tabletop workflow that turns real authority and contact references into retrievable, exercised preparation. Evidence: https://www.cisa.gov/audiences/small-and-medium-businesses · https://www.cisa.gov/stopransomware/ransomware-guide · https://www.ftc.gov/business-guidance/resources/data-breach-response-guide-business保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 82/100. Incident impact can be extreme and owners, MSPs, insurers and counsel form a clear buyer/partner set, but usage is low-frequency and free official guidance is abundant. Retain only because pre-incident authority, primary-identity-independent contacts, minimum service and tabletop evidence are distinct from the existing AI IncidentBrief response intake; no forensic or legal automation is promised.
#0023 AI 事件初报生成器Incident Brief
在线Live
69/100
中置信度medium confidence
18 / 13 / 8 / 17 / 7 / 4 / 2 一次性付费One-time payment
尚未形成成熟行为证据official-guidance
事件响应价值高且时效强,但购买频率低。PMF prior 69/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.事件价值高但低频;一次性响应包比会员更匹配。Retained for a time-boxed test; PMF has not been claimed.
#0003 AI 内容标识场景生成器AI Mark
在线Live
68/100
中置信度medium confidence
17 / 10 / 10 / 13 / 13 / 2 / 3 一次性付费One-time payment
分发前Pre-distribution
Article 50 带来短期分发窗口;免费标签和官方工具很多,差异化弱。PMF prior 68/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: pre_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.优先测试:法规时点带来分发窗口,但免费披露工具拥挤。Retained for a time-boxed test; PMF has not been claimed.
#0009 工具 Schema 重量分析器Schema Meter
在线Live
68/100
中置信度medium confidence
14 / 10 / 12 / 14 / 11 / 4 / 3 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据launched no distribution
工具 schema 成本可重复发生;已有免费计算器,自动化 API 才可能收费。PMF prior 68/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.优先测试:问题高频但竞品存在,以 CI/API 自动化区分。Retained for a time-boxed test; PMF has not been claimed.
#0045 紧急访问准备度预检Cloud Break Glass
在线Live
71/100
中置信度medium confidence
17 / 14 / 11 / 16 / 9 / 2 / 2 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 71/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Microsoft recommends two or more emergency identities, independent authentication, high-priority monitoring and validation at least every 90 days; AWS recommends predeployment and periodic testing before identity-service disruption; CISA recognizes tightly protected, multi-person emergency access. The job and cadence are explicit, while native guidance remains a strong free substitute. Evidence: https://learn.microsoft.com/en-us/entra/identity/role-based-access-control/security-emergency-access · https://learn.microsoft.com/en-us/entra/architecture/security-operations-privileged-accounts · https://docs.aws.amazon.com/singlesignon/latest/userguide/emergency-access.html · https://www.cisa.gov/sites/default/files/2023-05/tic_3.0_cloud_use_case_508c.pdf保留为限时验证;尚未声称已获得 PMF。Advance as a cross-platform drill-evidence workpaper, not an identity product. Quarterly validation creates repeat exposure and the buyer/risk are clear, but no form can verify privilege or custody, so one-time per-drill pricing and strict no-secret boundaries are appropriate.
#0117 处理最多 3 条虚构或脱敏 policy pathDependabot Cooldown Policy Map
在线Live
84/100
高置信度high confidence
17 / 13 / 14 / 17 / 13 / 7 / 3 按任务价值匹配one time dependency policy pack
尚未形成成熟行为证据launched no distribution
GitHub 当前一手公告确认,github.com 的 Dependabot 普通 version updates 现在默认等待新 release 在 registry 至少存在 3 天才开启 PR,而 security updates 仍立即开启;GHES 3.23 将获得相同行为。当前 options 文档提供 1–90 天、自定义 major/minor/patch、include/exclude 与 opt-out,其中 exclude 优先。真实治理任务是按 repository/ecosystem 复核新默认和显式覆盖,区分安全例外,并用生成 PR 与 CI 观察证明策略效果。 Evidence: https://github.blog/changelog/2026-07-14-dependabot-version-updates-introduce-default-package-cooldown/ · https://docs.github.com/en/code-security/reference/supply-chain-security/dependabot-options-referencePMF prior 84/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 84/100。默认变化覆盖大量仓库,version/security 分流和显式覆盖错误会造成过快引入或延迟修复,平台、安全与 repository owner 明确;但 Dependabot 原生 YAML/日志、Renovate、Snyk/Mend 和内部 dependency policy 是强替代。一次性 15 美元只在跨仓库证据交接中有有限价值,不能靠日期计算或安全叙事臆测需求。Retained for a time-boxed test; PMF has not been claimed.
#0024 AI 治理证据新鲜度检查Evidence Clock
在线Live
67/100
中低置信度medium-low confidence
16 / 11 / 13 / 14 / 7 / 3 / 3 会员订阅Membership
尚未形成成熟行为证据official-guidance
持续治理与会员逻辑一致;用户是否愿意外置月度复核未知。PMF prior 67/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.持续治理与会员匹配,但需要证明用户愿意每月回来。Retained for a time-boxed test; PMF has not been claimed.
#0010 中英文 AI 内容披露文案生成器Label Copy
在线Live
66/100
中置信度medium confidence
16 / 9 / 10 / 10 / 13 / 3 / 5 捐助Donation
尚未形成成熟行为证据launched no distribution
披露时点强且易做社交内容;免费直接替代丰富,只适合捐助型轻实验。PMF prior 66/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.限时测试:强时点和易分发,但直接免费替代很多。Retained for a time-boxed test; PMF has not been claimed.
#0104 Android 开发者验证交接图Mobile Developer Verification Map
在线Live
82/100
中低置信度medium-low confidence
18 / 13 / 12 / 18 / 13 / 5 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google current guidance links verified identity, account type, package name and signing-key ownership, with Play, non-Play, Console, Studio and API paths plus phased 2026/2027 enforcement. Native tools are strong substitutes; the testable wedge is cross-account release handoff. Evidence: https://developer.android.com/developer-verification/guides/android-developer-console · https://developer.android.com/developer-verification/guides/faq?hl=en · https://developer.android.com/developer-verification/guides/full-distribution保留为限时验证;尚未声称已获得 PMF。Package and signing-key responsibility can block distribution and creates a time-bound cross-team handoff; however Console, Studio and APIs already control actual state, so only a small one-time evidence-pack wedge deserves testing.
#0058 退款对账工具Cash Refund Proof
在线Live
81/100
中低置信度medium-low confidence
18 / 14 / 13 / 19 / 9 / 4 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 81/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Stripe and PayPal expose transaction-level payments, refunds, fees and payout identifiers, while current QuickBooks guidance still documents manual fees and deposits when a sales channel uses a third-party processor. The opportunity is only the cross-source gap among orders, processor balance activity and bank/accounting exports; native single-provider reporting is the strongest substitute. Evidence: https://docs.stripe.com/reports/payout-reconciliation · https://developer.paypal.com/docs/reports/reference/tcodes/ · https://quickbooks.intuit.com/learn-support/en-us/help-article/sales-logistics/record-third-party-sales-fees-quickbooks-online/L31nlfcQg_US_en_US保留为限时验证;尚未声称已获得 PMF。Pre-judgment 81/100: refund and fee mismatches create a clear month-close cost for bookkeepers and small ecommerce finance teams, and exact identifiers make part of the result auditable. Differentiation is narrow and fuzzy matches can harm trust, so this survives only as a local cross-source workpaper with exact-ID priority, human confirmation and measured repeat use.
#0120 在浏览器本地检查最多 3 条脱敏 audience-data 路径的 API/UI surface、identity use、access role、observed response、permission、aggregate/authorized/opt-in replacement、consent、retention、test、cutover、evidence 与 ownerGitHub Public Audience Access Migration Map
在线Live
84/100
中置信度medium confidence
18 / 14 / 13 / 18 / 13 / 5 / 3 按任务价值匹配one time audience access migration pack
尚未形成成熟行为证据launched no distribution
GitHub 当前公告确认 stargazers 与 watchers 身份列表将限制给管理员/协作者,users/{username}/subscriptions 在弃用期会返回空结果后移除,相关 API/UI 可能返回 empty 或 403,目标是减少 spam 和用户数据滥用。原生权限、aggregate counts、GitHub Apps、first-party analytics 与 opt-in 社群是强替代。 Evidence: https://github.blog/changelog/2026-06-30-upcoming-access-restrictions-to-public-api-endpoints-and-ui-views/ · https://docs.github.com/en/rest/activity/starring · https://docs.github.com/en/enterprise-cloud@latest/rest/activity/watching?apiVersion=2026-03-10PMF prior 84/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.84/100 PMF 预判:静默 200 空数组、403 和端点移除会破坏现有 DevRel/分析 pipeline,买家与隐私后果清晰;但客群窄且原生权限、聚合和自有同意数据替代强,只验证不绕过平台的访问迁移证据。Retained for a time-boxed test; PMF has not been claimed.
#0025 Agent 工具与网络 allowlist 生成器Tool Allowlist
在线Live
66/100
中置信度medium confidence
16 / 11 / 12 / 14 / 7 / 3 / 3 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据official-guidance
最小权限需求真实且可批量,但与 MCPPermit/安全平台重叠。PMF prior 66/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.真实自动化场景存在,但与 MCPPermit 重叠;排队验证。Retained for a time-boxed test; PMF has not been claimed.
#0060 预算与实际匹配工具Cash Budget Variance
在线Live
80/100
中低置信度medium-low confidence
18 / 14 / 13 / 18 / 9 / 4 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 80/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Xero directly offers a Budget Variance report; LivePlan connects plan and actual accounting data into a recurring performance dashboard and monthly review; Fathom combines Actual vs Budget, forecast snapshots and commentary and explicitly separates what happened, why and what it means. These prove the recurring buyer job while making basic variance arithmetic a free commodity. Evidence: https://central.xero.com/s/article/Budget-Variance-report · https://www.fathomhq.com/blog/how-to-explain-variance-to-board · https://www.liveplan.com/features/performance-dashboard保留为限时验证;尚未声称已获得 PMF。Pre-judgment 80/100: monthly cadence, finance ownership and decision impact are clear, but native accounting/reporting suites are strong and a number cannot prove its cause. The product survives only as a no-connection, source-key workpaper that keeps arithmetic separate from human evidence. $12 is one-time until next-month return proves subscription value.
#0044 备份隔离与恢复证据预检Cloud Backup Reach
在线Live
69/100
中低置信度medium-low confidence
18 / 13 / 10 / 16 / 8 / 2 / 2 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 69/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA recommends offline encrypted backups and regular recovery testing, while AWS, Azure and Google Cloud all provide native isolation, immutability or scheduled restore capabilities. The costly failure is real, but provider-native vaults are strong substitutes; the remaining wedge is a cloud-neutral evidence request that distinguishes configuration from an observed usable restore. Evidence: https://www.cisa.gov/stopransomware/ransomware-guide · https://docs.aws.amazon.com/aws-backup/latest/devguide/logicallyairgappedvault.html · https://docs.aws.amazon.com/aws-backup/latest/devguide/restore-testing.html · https://learn.microsoft.com/en-us/azure/backup/backup-azure-immutable-vault-how-to-manage · https://docs.cloud.google.com/backup-disaster-recovery/docs/concepts/backup-vault保留为限时验证;尚未声称已获得 PMF。Advance as a narrow one-scope recovery-evidence preflight. The buyer and downside are clear, yet usage is periodic and no questionnaire can prove recoverability. Price and claims therefore stay below audit/consulting positioning and require independent restore evidence.
#0059 折扣影响与交换条件工作表Deal Discount Gate
在线Live
80/100
中低置信度medium-low confidence
17 / 14 / 12 / 19 / 9 / 5 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 80/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Salesforce CPQ natively combines pricing, discount rules and automated cross-functional approval chains; HubSpot quote approvals currently route by discount, SKU, billing frequency, net terms and user attributes. These validate the job and create a strong substitute. The only wedge is a no-CRM small team that needs transparent contribution-profit and give-get math before adopting CPQ. Evidence: https://www.salesforce.com/sales/cpq/ · https://knowledge.hubspot.com/quotes/set-up-quote-approvals保留为限时验证;尚未声称已获得 PMF。Pre-judgment 80/100: the buyer, decision frequency and margin impact are clear, and the arithmetic is independently reproducible. Differentiation is narrow, so the product does not pretend to execute approval. The prior monthly model is rejected until repeat behavior exists; a $9 one-time workpack is the honest first revenue test.
#0006 AI 评测样本量规划器Eval Enough
在线Live
65/100
中置信度medium confidence
14 / 11 / 12 / 15 / 8 / 3 / 2 一次性付费One-time payment
分发前Pre-distribution
评测成本与决策价值存在;单个样本量计算器被免费统计工具替代。PMF prior 65/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: pre_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.限时测试:评测预算有经济价值,但需要团队工作流差异。Retained for a time-boxed test; PMF has not been claimed.
#0107 无需凭据的移动应用审核访问交接图Mobile Review Access Handoff
在线Live
82/100
中低置信度medium-low confidence
18 / 13 / 12 / 18 / 12 / 6 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google currently requires review sign-in details to remain valid, reusable, location-independent and complete for external identities and paywalls; Apple accepts review context per submission. Official store fields and secret managers are strong substitutes, leaving only credential-free coverage and owner handoff. Evidence: https://support.google.com/googleplay/android-developer/answer/15748846?hl=en-EN · https://developer.apple.com/help/app-store-connect/manage-submissions-to-app-review/overview-of-submitting-for-review · https://support.google.com/googleplay/android-developer/answer/2543765?hl=en保留为限时验证;尚未声称已获得 PMF。Review blockage has real release cost and crosses feature, identity, MFA, location, entitlement and on-call owners; however real secrets and submissions belong in native stores and approved secret systems, so only a credential-free one-time handoff deserves testing.
#0042 多供应商 API 回退预检API Fallback Route
在线Live
68/100
中低置信度medium-low confidence
17 / 13 / 9 / 14 / 9 / 4 / 2 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 68/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。AWS documents service-level failover controls and an explicit read-local/write-global data pattern; Azure shows that compensation is application-specific, fallible and often needs human intervention; Stripe demonstrates provider-specific idempotency behavior. These prove the job is real while also showing strong free and platform-native substitutes. Evidence: https://aws.amazon.com/blogs/compute/implementing-multi-region-failover-for-amazon-api-gateway/ · https://docs.aws.amazon.com/architecture-diagrams/latest/multi-region-api-gateway-with-cloudfront/multi-region-api-gateway-with-cloudfront.html · https://learn.microsoft.com/en-us/azure/architecture/patterns/compensating-transaction · https://learn.microsoft.com/en-us/azure/architecture/patterns/circuit-breaker · https://docs.stripe.com/api/idempotent_requests保留为限时验证;尚未声称已获得 PMF。Advance only as a narrow architecture preflight, not an automated failover product. Downtime and duplicate-write consequences create economic value, but the task is episodic, production evidence is indispensable and vendor-native guidance limits differentiation.
#0026 Agent 红队范围规划器Red Team Plan
在线Live
65/100
中低置信度medium-low confidence
17 / 12 / 7 / 17 / 6 / 4 / 2 按任务价值匹配qualified lead
尚未形成成熟行为证据official-guidance
专业服务价值高,但工具本身低频且依赖合作方。PMF prior 65/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.专业服务客单高但低频且需合作方;先验证合格线索。Retained for a time-boxed test; PMF has not been claimed.
#0086 的浏览器本地 2026 年标准里程法与实际费用法证据准备度比较Books Mileage Method Compare
在线Live
81/100
中置信度medium confidence
17 / 14 / 9 / 18 / 14 / 5 / 4 按任务价值匹配disclosed affiliate referral
尚未形成成熟行为证据launched no distribution
PMF 先验 81/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。IRS primary sources confirm a mid-year 2026 business mileage-rate change and durable method-choice and recordkeeping constraints. The buyer and exact-intent query are clear, but annual frequency, free official information and mature mileage/accounting substitutes constrain retention and differentiation; this is a high prior for completion intent, not yet proof of recurring revenue. Evidence: https://www.irs.gov/tax-professionals/standard-mileage-rates · https://www.irs.gov/publications/p463 · https://www.irs.gov/irb/2026-29_irb保留为限时验证;尚未声称已获得 PMF。PMF prior 81/100: a dated 2026 rate split creates urgent exact-intent discovery and the record-evidence gap has economic consequences for owners and bookkeepers. Recurrence is weak and substitutes are strong, so the complete tool stays free; monetization is credible only if users complete comparisons and then choose a clearly disclosed relevant partner.
#0126 在浏览器本地检查最多 3 条脱敏 cohort 的 enablement、active/unique committer band、AI/Actions cost surface、existing-tool overlap、baseline、gate、remediation、cost cap、quality/defect outcome 与 ownerGitHub Code Quality Rollout Economics Map
在线Live
85/100
中置信度medium confidence
18 / 14 / 13 / 18 / 13 / 6 / 3 按任务价值匹配one time code quality rollout pack
尚未形成成熟行为证据launched no distribution
GitHub 官方资料确认 Code Quality 于 2026-07-20 GA 并开始收费,组织级 repository access 可批量启停;成本来自每名 active committer 每月 10 美元、AI credits 与 Actions compute,ruleset evaluate mode 支持分阶段试点。需求是启用范围、总成本与增量质量结果的跨 owner 决策,不是又一个质量分数。 Evidence: https://github.blog/changelog/2026-07-20-github-code-quality-is-now-generally-available/ · https://docs.github.com/en/billing/concepts/product-billing/github-code-quality · https://docs.github.com/en/code-security/concepts/about-code-qualityPMF prior 85/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.85/100:自动计费、三类成本和组织级批量启用形成明确 DevEx/FinOps 决策;但原生 dashboard/billing-impact dialog、linters、SonarQube/Code Climate 与表格替代强,所以只验证一次性 cohort rollout economics 证据包。Retained for a time-boxed test; PMF has not been claimed.
#0061 AI 治理变更登记Gov Board Packet
在线Live
79/100
中低置信度medium-low confidence
17 / 14 / 12 / 18 / 9 / 4 / 5 按任务价值匹配prepaid report
尚未形成成熟行为证据launched no distribution
PMF 先验 79/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。AI Governance Hub already sells uploaded-data-to-board-ready PDF, DOCX and PPTX reporting; Veytrio markets executive risk summaries, monthly trends and board packs; Credo AI covers the broader governance lifecycle. These establish a recurring buyer and paid reporting job, but also show that the proposed output is directly substituted and differentiation is weak. Evidence: https://www.aigovernancehub.ai/ · https://www.veytrio.com/compliance · https://www.credo.ai/product保留为限时验证;尚未声称已获得 PMF。Pre-judgment 79/100: governance leaders and CISOs have a recurring, board-visible reporting job with credible economic value, but direct report generators and enterprise governance platforms already serve it. Launch only a redacted structured-draft wedge; require payment and same-team repeat before increasing investment.
#0101 Apple 隐私清单组件交接图Mobile Privacy Manifest Map
在线Live
81/100
高置信度high confidence
17 / 13 / 11 / 18 / 12 / 7 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
PMF 先验 81/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Apple currently requires privacy manifests and approved truthful reasons for covered API use, and treats App and third-party SDK components separately. The release risk and accountable mobile buyer are real, but Xcode, App Store Connect and SDK-vendor artifacts are strong native substitutes. The independent wedge is a sanitized cross-component responsibility and evidence handoff before the exact archive is checked. Evidence: https://developer.apple.com/documentation/bundleresources/privacy-manifest-files · https://developer.apple.com/documentation/bundleresources/describing-use-of-required-reason-api · https://developer.apple.com/documentation/bundleresources/adding-a-privacy-manifest-to-your-app-or-third-party-sdk保留为限时验证;尚未声称已获得 PMF。PMF prior 81/100: advance as a narrow one-time release evidence pack, not a scanner or compliance product. A $12 price can match the avoided cost of one blocked submission or coordination gap, while the manual cross-component register differentiates only when teams find native reports fragmented. Do not infer demand from launch or zero traffic; behavior and payment must meet the fixed gates.
#0014 AI 供应商 RFP 范围生成器Stack RFP
在线Live
64/100
低置信度low confidence
13 / 12 / 8 / 16 / 8 / 4 / 3 按任务价值匹配paid listing
尚未形成成熟行为证据launched no distribution
供应商获客可收费,但没有买方流量前付费收录是弱假设。PMF prior 64/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.限时测试:供应商愿意获客,但需先建立买方流量。Retained for a time-boxed test; PMF has not been claimed.
#0039 Agent 工具预授权影响范围图Agent Tool Blast
在线Live
67/100
中低置信度medium-low confidence
17 / 12 / 9 / 14 / 10 / 3 / 2 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 67/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。OWASP and Microsoft support the underlying least-privilege, tool-misuse and blast-radius problem. Anthropic already exposes approval policy controls, Microsoft has enterprise posture assessment, and a free MCP static blast-radius auditor is a direct substitute. The remaining wedge is a no-code, cross-framework design review before authorization. Evidence: https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/ · https://learn.microsoft.com/en-us/security/zero-trust/sfi/manage-agentic-risk · https://platform.claude.com/docs/en/managed-agents/permission-policies · https://pypi.org/project/mcp-blast-radius/保留为限时验证;尚未声称已获得 PMF。67/100 test, not PMF: real and consequential buyer problem with reachable developer channels, but low differentiation and strong free/platform substitutes. Continue only after 21 days and 150 qualified visits if completion reaches 10% and at least one genuine $11 purchase or paid intent reaches 3%; otherwise stop at the preset sample gate.
#0027 评测刷新频率规划器Eval Drift
在线Live
64/100
中低置信度medium-low confidence
15 / 10 / 12 / 13 / 7 / 4 / 3 会员订阅Membership
尚未形成成熟行为证据official-guidance
持续评测真实,但现有平台与内部流程替代强。PMF prior 64/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: official-guidance. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.持续问题存在,但免费评测平台替代强;严格限时测试会员。Retained for a time-boxed test; PMF has not been claimed.
#0007 AI Batch 决策器Batch Or Now
已关闭Closed
63/100
中置信度medium confidence
15 / 9 / 11 / 16 / 8 / 2 / 2 一次性付费One-time payment
组合先验关闭Closed by portfolio prior
节省直接可量化;官方折扣明确且内部脚本易替代,独立付费弱。PMF prior 63/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:官方 Batch API 与内部脚本替代强,一次性工具难形成稳定月收入。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0043 三方 CSV 匹配工具Cash Invoice Match
在线Live
67/100
中低置信度medium-low confidence
17 / 13 / 13 / 13 / 7 / 2 / 2 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据launched no distribution
PMF 先验 67/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。QuickBooks already suggests and confirms bank-to-invoice/payment matches, and Stripe provides processor-native bank reconciliation for eligible accounts. That makes native substitutes strong. The remaining narrow job is a recurring, cross-system CSV handoff where invoices, processor settlements and bank rows are not in one connected ledger. Evidence: https://quickbooks.intuit.com/learn-support/en-us/help-article/bank-transactions/match-transactions-quickbooks-online/L0MF3Fn6y_US_en_US · https://quickbooks.intuit.com/learn-support/en-us/help-article/statement-reconciliation/reconcile-account-quickbooks-online/L3XzsllsK_US_en_US · https://docs.stripe.com/bank-reconciliation · https://www.irs.gov/pub/irs-pdf/p583.pdf保留为限时验证;尚未声称已获得 PMF。Advance as a constrained cross-system CSV experiment, not an accounting platform. Monthly close frequency and a clear bookkeeper/operator buyer support prepaid usage, but one-to-one matching misses split, batch, fee, FX and dispute cases, so the score remains conservative.
#0063 每周服务台流程与 SLA 复核Desk Desk Memo
在线Live
77/100
中低置信度medium-low confidence
16 / 14 / 14 / 15 / 10 / 4 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 77/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Atlassian Jira Service Management supplies native and custom ITSM reports and granular CSV export paths; Zendesk Explore supplies prebuilt and customizable shared dashboards. These validate a frequent reporting job while showing that a metrics-only summary is heavily substituted. The only independent hypothesis is an explicit weekly decision, owner, baseline, target and next-week outcome loop. Evidence: https://support.atlassian.com/jira-service-management-cloud/docs/create-or-edit-an-itsm-report/ · https://support.atlassian.com/jira/kb/export-report-data-from-a-jsm-project-using-work-item-filters/ · https://support.zendesk.com/hc/en-us/articles/4408846844826-Understanding-Explore-dashboards保留为限时验证;尚未声称已获得 PMF。Pre-judgment 77/100: service-desk managers have a weekly operating cadence and measurable backlog/SLA economics, but native ITSM analytics make reporting itself commodity. Launch only the browser-local decision-loop wedge; membership survives only if teams return four weeks, review prior decisions and pay, not because a CSV can be summarized.
#0088 双地点公开事实证据矩阵Discover Location Page Evidence
在线Live
80/100
中低置信度medium-low confidence
17 / 13 / 10 / 16 / 14 / 6 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 80/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google primary policy identifies doorway abuse involving regional or city pages that funnel users and substantially similar pages, but publishes no mandatory uniqueness percentage. Ten independently reviewed public location-page samples show a real operational need to separate shared facts from location-specific hours, services, facilities and restrictions. Mature SEO suites and agency audits are strong substitutes, so the defendable wedge is a manual public-fact receipt matrix, not duplicate-content scoring or copy generation. Evidence: https://developers.google.com/search/docs/essentials/spam-policies · https://developers.google.com/search/docs/appearance/structured-data/local-business · https://www.wholefoodsmarket.com/stores/springlakewall · https://www.mayoclinic.org/patient-visitor-guide/arizona/campus-buildings-maps/mayo-clinic-hospital保留为限时验证;尚未声称已获得 PMF。PMF prior 80/100: multi-location brands and agencies face a real, economically relevant page-review job and exact SEO/LBS search intent. The project is episodic and the audit market is crowded; a $12 one-time 3–20 location evidence workpack fits better than membership. Continue only if users enter real multi-location evidence and value owner/recheck structure instead of asking for AI city-name pages or a similarity score.
#0100 在浏览器本地复核一份脱敏模板的 HTTPS List-Unsubscribe、精确 Post 值与声明的 DKIM h= 覆盖Mail One Click Unsubscribe
在线Live
80/100
中低置信度medium-low confidence
17 / 13 / 11 / 17 / 13 / 6 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
RFC 8058 明确要求一个 HTTPS URI、精确的 List-Unsubscribe-Post 值、覆盖两个字段的有效 DKIM 签名和无重定向 POST;Gmail 当前批量发件要求又形成明确发布触发。任务真实,但 RFC、Gmail 文档和 ESP 原生设置均是强免费替代,独立产品只剩静态字段与人工运行证据分离的窄缝。 Evidence: https://www.rfc-editor.org/rfc/rfc8058.html · https://support.google.com/mail/answer/81126?hl=en · https://support.google.com/mail/answer/14229414?hl=enPMF prior 80/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 先验 80/100:发件负责人、发送前触发、投递与用户选择风险及精确搜索意图均明确;但 ESP 原生能力很强,语法检查极易商品化。只允许用一个免费模板到 7 美元多模板证据包验证,绝不把规范要求或页面访问当作 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0062 单条产品洞察适用性复核Product Evidence Age
在线Live
77/100
中低置信度medium-low confidence
16 / 13 / 11 / 17 / 10 / 5 / 5 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 77/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Dovetail explicitly describes repository decay, accountable maintenance, periodic audits and retirement of outdated insights, validating the problem. Dovetail workspace fields, Condens artifact fields and Productboard insight-to-feature links are strong current substitutes, so an independent tool has only a narrow no-migration applicability-review wedge and weak differentiation. Evidence: https://dovetail.com/research/synthesize-research-across-product-squads-unified-insight-library/ · https://docs.dovetail.com/academy/intro-to-workspace-fields · https://condens.io/help/using-condens/research-repository-functionality/artifact-information/ · https://support.productboard.com/hc/en-us/articles/10071375851155-Support-your-feature-ideas-with-customer-insights保留为限时验证;尚未声称已获得 PMF。Pre-judgment 77/100: stale research can misdirect active product decisions and the buyer is identifiable, but full research and product platforms already store metadata and linked insights. Launch only a browser-local one-record review; never infer truth from age, and require real status changes, payment and later return before investing.
#0087 账户删除发布准备度工具Trust Account Delete Brief
在线Live
80/100
中低置信度medium-low confidence
18 / 13 / 9 / 17 / 12 / 6 / 5 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 80/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current Apple and Google primary guidance creates a concrete store-release trigger: qualifying apps need clear account-deletion paths, associated-data handling and user communication, while Google also requires an external web resource. The accountable developer buyer and exact-intent discovery route are clear; official checklists, internal spreadsheets and broad DSAR platforms remain strong substitutes, and episodic releases limit recurrence. Evidence: https://developer.apple.com/support/offering-account-deletion-in-your-app/ · https://support.google.com/googleplay/android-developer/answer/13327111?hl=en · https://www.ftc.gov/business-guidance/resources/protecting-personal-information-guide-business保留为限时验证;尚未声称已获得 PMF。PMF prior 80/100: the release gate, buyer and consequence are real, and a local dependency/test map is narrower and safer than a DSAR platform. Frequency is episodic and the documents themselves are free, so the fit is a $12 one-time workpack—not membership—and survival requires completed real multi-system maps plus purchase intent after the age/sample gate.
#0041 OpenAPI OAuth 权限范围对比工具API Auth Scope
在线Live
66/100
中低置信度medium-low confidence
16 / 12 / 10 / 13 / 10 / 3 / 2 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 66/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。OAuth scope selection and OpenAPI security declarations create a real design-review job, but GitHub, Google, AWS and other providers already supply strong native permission guidance or analysis. The plausible wedge is a token-free, cross-provider contract comparison for small API teams; it is not an enforcement or IAM verdict. Evidence: https://www.rfc-editor.org/info/rfc6749/ · https://spec.openapis.org/oas/latest.html · https://docs.github.com/en/rest/authentication/permissions-required-for-fine-grained-personal-access-tokens · https://developers.google.com/identity/protocols/oauth2/production-readiness/sensitive-scope-verification · https://docs.aws.amazon.com/access-analyzer/latest/APIReference/Welcome.html保留为限时验证;尚未声称已获得 PMF。Advance as a narrow paid experiment because the buyer can finish a useful local comparison without giving us a token. Keep the score conservative because provider-native tools are stronger and the likely usage frequency is episodic.
#0028 AI 供应商退出准备度Vendor Exit
已关闭Closed
63/100
中低置信度medium-low confidence
16 / 12 / 7 / 16 / 5 / 5 / 2 按任务价值匹配qualified lead
组合先验关闭Closed by portfolio prior
退出价值高但事件频率低、交付复杂。PMF prior 63/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:问题高价值但极低频,且依赖尚未建立的迁移服务伙伴。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0015 AI 评测 Rubric 检查器Eval Rubric
已关闭Closed
62/100
低置信度low confidence
13 / 9 / 11 / 13 / 9 / 4 / 3 广告Advertising
组合先验关闭Closed by portfolio prior
教育内容和搜索意图存在;广告收入需要显著规模,先验证内容传播。PMF prior 62/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:广告收入需要远超当前可分配流量的规模。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0040 交易证据分流工具Cash Audit Bundle
在线Live
65/100
中低置信度medium-low confidence
16 / 12 / 8 / 13 / 9 / 4 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 65/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。IRS guidance supports the need to retain books and supporting transaction documents, while QuickBooks and Xero already provide strong in-platform audit histories and professional audit tools cover formal sampling. The remaining testable wedge is a cross-system, browser-local evidence-gap queue that never claims statistical representativeness or record sufficiency. Evidence: https://www.irs.gov/pub/irs-pdf/p583.pdf · https://www.irs.gov/businesses/small-businesses-self-employed/how-long-should-i-keep-records · https://quickbooks.intuit.com/learn-support/en-uk/help-article/audit-log/use-audit-log-quickbooks-online/L2WoVnW6I_GB_en_GB · https://central.xero.com/s/article/View-a-history-and-notes-summary-for-transactions-and-user-activity · https://www.aicpa-cima.com/cpe-learning/publication/audit-sampling-audit-guide保留为限时验证;尚未声称已获得 PMF。65/100 test, not PMF: the bookkeeping evidence problem and periodic trigger are real, and $11 fits a one-off close or evidence-prep task, but accounting platforms, accountants and audit software are strong substitutes. Continue only at the preset age/sample gate with real local CSV use and genuine payment evidence.
#0029 AI 成本 SLO 规划器Cost SLO
已关闭Closed
61/100
低置信度low confidence
15 / 10 / 12 / 14 / 5 / 2 / 3 会员订阅Membership
组合先验关闭Closed by portfolio prior
成本控制高频,但供应商账单和 OTel 免费替代强。PMF prior 61/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:免费账单、FinOps 与 OTel 替代强,会员模式不匹配。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0070 版在浏览器本地接受最多 5 条抽象 allow/deny 边界规则Agent Context Leak
在线Live
76/100
中置信度medium confidence
18 / 13 / 12 / 17 / 9 / 5 / 2 预充值 API 用量Prepaid API usage
尚未形成成熟行为证据launched no distribution
OWASP 将身份、工具、记忆、目标与自主动作列为 Agent 安全风险;OpenAI Agents SDK 说明本地上下文可被嵌套 Agent 共享且记忆隔离依赖布局,AWS AgentCore 提供运行时会话隔离。它们证明验证任务与强运行时替代同时存在,但不证明用户会为独立测试设计器付款。 Evidence: https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/ · https://openai.github.io/openai-agents-python/context/ · https://openai.github.io/openai-agents-python/sandbox/memory/ · https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-sessions.htmlPMF prior 76/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 76/100:Agent 平台与安全工程买家清楚,跨租户/任务/工具泄漏后果高且会重复测试;差异化来自供应商无关的抽象测试矩阵,而非运行时保护。Promptfoo 类红队工具、框架原生隔离和内部 harness 都是强替代,必须用批量需求、API 意向和购买验证。Retained for a time-boxed test; PMF has not been claimed.
#0089 人工注册与取消路径记录器Billing Cancel Path Audit
在线Live
79/100
中低置信度medium-low confidence
18 / 13 / 10 / 18 / 11 / 5 / 4 按任务价值匹配one time audit pack
尚未形成成熟行为证据launched no distribution
PMF 先验 79/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。FTC current primary materials describe continuing negative-option complaints, difficult cancellation concerns and a 2026 advance notice seeking comment; they also make clear that the 2024 rule was vacated, so no obsolete click-to-cancel claim is used. Stripe documents a mature built-in self-service cancellation and retention substitute. The buyer and economic risk are real, but usability tests, billing portals, churn tools and legal review create a thin, episodic wedge. Evidence: https://www.ftc.gov/news-events/news/press-releases/2026/03/ftc-seeks-public-comment-response-advance-notice-proposed-rulemaking-regarding-negative-option · https://www.ftc.gov/legal-library/browse/rules/negative-option-rule · https://docs.stripe.com/customer-management/cancellation-page保留为限时验证;尚未声称已获得 PMF。PMF prior 79/100: an independent signup-versus-cancellation evidence comparison can expose operational friction without becoming a retention executor, and the SaaS product/operations buyer is reachable. Differentiation and recurrence are weak, so $12 once fits better than subscription. This remains viable only if users complete both paths and value the evidence/retest artifact—not if they only need Stripe/Churnkey execution or legal advice.
#0064 续约证据整理器Deal Renewal Signal
在线Live
76/100
中低置信度medium-low confidence
17 / 14 / 12 / 19 / 8 / 3 / 3 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 76/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Salesforce explicitly aggregates adoption, support and relationship signals for renewal likelihood; Gainsight centralizes health, product usage, support history, renewal timelines and stakeholders; HubSpot provides a customer-success workspace, tasks, alerts, reports and configurable health scores. These prove the renewal job and budget while making a standalone score or dashboard non-differentiated. Evidence: https://help.salesforce.com/s/articleView?id=analytics.csi_score_work_metrics_chs.htm&language=en_US&type=5 · https://www.gainsight.com/customer-success/ · https://knowledge.hubspot.com/customer-success/use-the-customer-success-workspace保留为限时验证;尚未声称已获得 PMF。Pre-judgment 76/100: renewal evidence has clear economic value and an identifiable small-team buyer, but full CS/CRM platforms dominate. Test only a one-time no-integration pack that separates dated facts, customer-confirmed outcomes, missing proof and assumptions. Stop if buyers need real-time integration or predictive scoring.
#0065 最小设备、软件、数据和外部服务登记Secure Asset List
在线Live
76/100
中低置信度medium-low confidence
18 / 13 / 13 / 15 / 10 / 3 / 4 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 76/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。NIST and CIS establish a basic asset inventory as a real security prerequisite. Snipe-IT is a strong free/open-source substitute and Intune automates application inventory, so the viable wedge is only a lightweight browser-local prep register for very small teams that have neither ITAM nor MDM. Evidence: https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.1300.pdf · https://www.cisecurity.org/controls/inventory-and-control-of-enterprise-assets · https://snipeitapp.com/ · https://learn.microsoft.com/en-us/intune/app-management/deployment/enhanced-app-inventory保留为限时验证;尚未声称已获得 PMF。Launch the narrow manual questionnaire and insurance-prep wedge, not an asset-discovery product. Stop or redirect users to established ITAM/MDM tools if evidence shows they need automatic scanning, authoritative completeness, or ongoing fleet management.
#0099 在浏览器本地完成最多三条抽象 Passkey 注册、登录、管理、恢复、重新授权或回退场景的预期与人工观察证据检查Secure Passkey Recovery
在线Live
79/100
中低置信度medium-low confidence
18 / 13 / 10 / 18 / 11 / 6 / 3 按任务价值匹配one time release evidence pack
尚未形成成熟行为证据launched no distribution
截至 2026-07-27,NIST、FIDO、Google 与 Apple 的当前一手资料共同证明 Passkey 上线不仅是注册按钮,还涉及登录、凭据管理、跨设备、删除、恢复、重新授权和回退;Auth0 等 IdP 已原生覆盖实现,因此独立产品只保留跨供应商发布证据收据这一窄缝。 Evidence: https://pages.nist.gov/800-63-4/sp800-63b.html · https://fidoalliance.org/implement-passkeys-overview/ · https://developers.google.com/identity/passkeys/ux/user-journeys · https://developer.apple.com/documentation/AuthenticationServices/supporting-passkeys · https://auth0.com/docs/authenticate/database-connections/passkeysPMF prior 79/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 79/100:Passkey 迁移风险和跨团队发布交接具有真实经济后果,但原生 IdP、FIDO 指南与内部测试是强替代。只验证一次性 $9 证据包;不把标准热度、验证 checkout 或零流量当 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0031 本地模型 vs 托管 API 路线规划器Local Or API
已关闭Closed
59/100
低置信度low confidence
14 / 8 / 8 / 13 / 7 / 6 / 3 联盟分成Affiliate commission
组合先验关闭Closed by portfolio prior
搜索面广但买家模糊、比较内容拥挤且快速过时。PMF prior 59/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:买家过宽、维护重,且没有正式联盟收入路径。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0090 六阶段表单错误恢复证据记录器Access Error Message Route
在线Live
78/100
中置信度medium confidence
18 / 12 / 10 / 17 / 12 / 6 / 3 按任务价值匹配one time regression pack
尚未形成成熟行为证据launched no distribution
PMF 先验 78/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。W3C primary guidance establishes that detected input errors need textual identification and that complete accessibility evaluation requires human judgment. Ten distinct public example shapes were manually mapped into boundary fixtures; mature scanners and expert audits remain strong substitutes, so the opportunity is a narrow release-evidence workflow rather than another scanner. Evidence: https://www.w3.org/WAI/WCAG22/Understanding/error-identification.html · https://www.w3.org/WAI/tutorials/forms/validation/ · https://www.w3.org/WAI/test-evaluate/ · https://www.w3.org/TR/WCAG22/保留为限时验证;尚未声称已获得 PMF。PMF prior 78/100: form-error regressions can block releases and expose costly remediation, with identifiable accessibility QA, product and engineering buyers. The manual six-phase receipt is useful, but scanner suites and specialist audits dominate automated detection. Survival depends on buyers paying for compact, repeatable regression evidence—not on a compliance promise.
#0002 提示缓存盈亏计算器Cache Check
已关闭Closed
56/100
中置信度medium confidence
12 / 8 / 10 / 12 / 8 / 3 / 3 一次性付费One-time payment
组合先验关闭Closed by portfolio prior
缓存节省真实但免费计算器众多;只有真实工作负载集成才有持续价值。PMF prior 56/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.组合后 30%先验停止:免费计算器和供应商原生指标替代强,一次性补丁包复用弱。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0127 在浏览器本地检查最多 3 条脱敏 review-control 路径的 cohort/target、ruleset/protection layer、dismissal actor class、bypass、stale/code-owner review、emergency、allow/deny tests、Rule Insights/audit、rollback 和 ownerGitHub Review Dismissal Authority Map
在线Live
83/100
中低置信度medium-low confidence
17 / 14 / 12 / 18 / 13 / 6 / 3 按任务价值匹配one time review authority pack
尚未形成成熟行为证据launched no distribution
GitHub 官方公告确认 repository rulesets 已可原生选择 users、teams 与 apps 限制谁能清除 pull-request review;官方规则文档确认多个 rulesets 可叠加并与 classic branch protection 共存,REST schema 暴露 dismissal_restriction。需求是跨仓库有效权限、例外和 allow/deny 结果的治理证据,不是复制 actor picker。 Evidence: https://github.blog/changelog/2026-07-07-restrict-who-can-dismiss-reviews-in-rulesets/ · https://docs.github.com/en/repositories/configuring-branches-and-merges-in-your-repository/managing-rulesets/about-rulesets · https://docs.github.com/en/rest/orgs/rulesPMF prior 83/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.83/100:清除审批权限直接影响 merge governance,新原生控制与多层规则带来明确迁移和测试任务;但 ruleset UI/API、Rule Insights、Terraform/settings managers 与 policy-as-code 替代很强,只验证一次性 authority layering 证据包。Retained for a time-boxed test; PMF has not been claimed.
#0066 功能全生命周期成本区间工具Product Feature Cost
在线Live
75/100
中低置信度medium-low confidence
16 / 13 / 11 / 16 / 10 / 4 / 5 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 75/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Jira Product Discovery already supports formula-based impact and effort decisions, Linear covers delivery estimates, and Swarmia now translates engineering investment into monetary cost. These substitutes validate the buyer problem but sharply limit differentiation; the remaining wedge is a private, sourced lifecycle band spanning maintenance, support, infrastructure, vendors, retirement and later actual-cost calibration. Evidence: https://support.atlassian.com/jira-product-discovery/docs/expression-based-formulas/ · https://www.atlassian.com/software/jira/product-discovery/pricing · https://linear.app/docs/estimates · https://www.swarmia.com/changelog/2026-06-16-fte-cost/保留为限时验证;尚未声称已获得 PMF。Launch only the narrow browser-local lifecycle cost-band and calibration worksheet. It must not become a project manager or roadmap score. Continue only if teams return with actuals and can name decisions changed; otherwise strong native substitutes make the standalone product unviable.
#0068 版在浏览器本地生成一个比率指标合同Product Metric Contract
在线Live
75/100
中置信度medium confidence
16 / 13 / 11 / 15 / 10 / 5 / 5 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
Amplitude、Eppo 与 Statsig 的官方文档共同证明:指标定义、认证指标和比率指标是产品与数据团队的真实工作对象;但成熟分析平台已覆盖计算层,因此本产品只把“定义冲突与审批证据包”作为更窄的购买假设,不能把问题存在误写成付款证据。 Evidence: https://amplitude.com/docs/analytics/metrics · https://docs.geteppo.com/data-management/certified-metrics/eppo-schema/ · https://docs.geteppo.com/data-management/metrics/ratio-metric/ · https://docs.statsig.com/statsig-warehouse-native/metrics/ratioPMF prior 75/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 为 75/100:买家和重复审阅场景清晰,错误口径会直接影响发布与经营决策;但替代品强、单纯模板容易被复制。只有跨产品与数据团队反复导入、版本回访并产生付款,才证明独立产品而不是一次性文档。Retained for a time-boxed test; PMF has not been claimed.
#0005 Agent 重试预算器Retry Budget
已关闭Closed
42/100
中置信度medium confidence
10 / 5 / 7 / 8 / 6 / 2 / 4 一次性付费One-time payment
组合先验关闭Closed by portfolio prior
问题由免费 SRE 文档和框架默认策略覆盖;独立使用与买家都弱。PMF prior 42/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.停止分发:工程常识和免费文档覆盖,独立计算器买家弱。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0091 有序百分比与固定金额订阅折扣场景Billing Discount Stack Guard
在线Live
77/100
中低置信度medium-low confidence
17 / 12 / 10 / 18 / 10 / 6 / 4 按任务价值匹配one time scenario pack
尚未形成成熟行为证据launched no distribution
PMF 先验 77/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current Stripe documentation supports up to 20 subscription discounts, defines once/repeating/forever duration behavior, and says mixed fixed/percentage order changes results. Stripe invoice preview is the authoritative execution substitute. Shopify native combinations and the current StackGuard checkout app show strong platform-level competition; RushOS therefore occupies only a narrow synthetic preflight and reconciliation wedge. Evidence: https://docs.stripe.com/billing/subscriptions/coupons · https://docs.stripe.com/invoicing/preview · https://help.shopify.com/en/manual/discounts/discount-combinations · https://apps.shopify.com/stackguard-app保留为限时验证;尚未声称已获得 PMF。PMF prior 77/100: a wrong recurring discount stack can damage contribution on every renewal, and billing, revenue operations and finance ownership are identifiable. The simple ordered arithmetic is understandable and searchable, but free calculators and billing-platform previews make differentiation thin. Survival requires repeated multi-plan comparison or a $9 decision-pack purchase, not calculator page views.
#0001 AI 应用真实利润闸门Margin Gate
已关闭Closed
39/100
中置信度medium confidence
9 / 5 / 6 / 7 / 7 / 1 / 4 一次性付费One-time payment
组合先验关闭Closed by portfolio prior
通用计算器拥挤、使用低频、无持续工作流,$1 静态包经济性也差。PMF prior 39/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: portfolio_prior_close. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.停止分发:通用利润计算器低频、竞争密集,缺少持续付费工作流。Structurally closed after the lower portfolio ranking; the reason remains in the public review.
#0072 版在浏览器本地建立一份 T-90 至 T+30 风险矩阵Product Sunset Plan
在线Live
75/100
中低置信度medium-low confidence
16 / 13 / 8 / 18 / 10 / 6 / 4 按任务价值匹配one time workpack
尚未形成成熟行为证据launched no distribution
LaunchDarkly 的官方文档证明成熟平台已管理旗标引用、评估状态、依赖与生命周期;Atlassian 的公开弃用政策和 Stripe 的版本规则证明外部通知、期限与迁移承诺需要正式流程。它们同时表明技术工具和公开政策是强替代,独立产品只能靠跨职能遗漏发现而成立。 Evidence: https://launchdarkly.com/docs/guides/flags/technical-debt · https://launchdarkly.com/docs/home/flags/flag-lifecycle-settings · https://developer.atlassian.com/cloud/commerce-it-frontoffice/platform-concepts/deprecation-policy/ · https://docs.stripe.com/sdks/versioningPMF prior 75/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 75/100:产品/工程买家明确,错误下线会破坏客户工作流并触发合同、数据与支持成本;但事件低频,模板和项目管理工具足以覆盖许多团队。真实停止日期、跨职能审阅、可行动遗漏与 49 美元购买必须共同验证,不把风险严重性等同于 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0074 整数分捆绑贡献毛利引擎Shop Bundle Margin
在线Live
75/100
中低置信度medium-low confidence
16 / 13 / 10 / 17 / 10 / 5 / 4 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
PMF 先验 75/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Shopify documents native bundle construction and profit reports, so substitutes are strong. Its own AOV guidance warns that a larger order does not automatically produce more profit. The narrow opportunity is a pre-publish, source-labeled, integer-cent contribution-margin stress test that keeps bundle component, fee, fulfillment, return and acquisition assumptions visible before a merchant changes the storefront. Evidence: https://help.shopify.com/en/manual/products/bundles/shopify-bundles · https://help.shopify.com/en/manual/products/bundles/eligibility-and-considerations · https://help.shopify.com/en/manual/reports-and-analytics/shopify-reports/report-types/default-reports/profit-reports · https://www.shopify.com/blog/average-order-value保留为限时验证;尚未声称已获得 PMF。PMF prior is 75/100: merchants have a recurring and economically meaningful pricing decision, and a one-cent error is testable. Buyer clarity is good for small Shopify merchants, but Shopify reports and spreadsheets cover much of the job, distribution is crowded, and a calculator alone has a shallow moat. Launch is justified as a narrow decision wedge, not as an accounting or AOV-growth product.
#0067 发布前决策合同Product Launch Gate
在线Live
74/100
中低置信度medium-low confidence
16 / 13 / 10 / 16 / 10 / 4 / 5 一次性付费One-time payment
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。LaunchDarkly already provides review approvals, guarded rollouts, metric monitoring and rollback, while Statsig attaches safeguards and health checks to releases. These platforms prove that release risk has budget, but they also make a generic readiness checklist weak. The remaining wedge is a low-friction, pre-platform value–risk–observation contract for teams that have not yet connected execution tooling. Evidence: https://launchdarkly.com/docs/home/releases/approvals/ · https://launchdarkly.com/docs/home/releases/managing-guarded-rollouts · https://docs.statsig.com/feature-flags/safeguards-overview · https://www.statsig.com/pricing保留为限时验证;尚未声称已获得 PMF。Launch only the browser-local pre-release contract and explicit contradiction check. It cannot call a release ready, deploy, monitor or approve. The product survives only if cross-functional review discovers gaps that actually delay or change launches and produces unrelated paid packs.
#0122 在浏览器本地映射最多 3 条脱敏更新路径的声明/实际 Python、manifest/lock、Dependabot job、支持目标、依赖测试、CI、新 PR 恢复、fallback、rollback、证据和 ownerDependabot Python39 Update Recovery Map
在线Live
82/100
高置信度high confidence
17 / 13 / 13 / 18 / 12 / 6 / 3 按任务价值匹配one time dependency recovery pack
尚未形成成熟行为证据launched no distribution
GitHub 已公告自 2026-06-23 起 Dependabot 不再支持已 EOL 的 Python 3.9,继续依赖该运行时可能不再收到依赖更新 PR;官方配置文档和 Python 版本状态表共同界定真实迁移与验证边界。 Evidence: https://github.blog/changelog/2026-05-19-upcoming-deprecation-of-python-3-9-for-dependabot/ · https://docs.github.com/en/code-security/how-tos/secure-your-supply-chain/secure-your-dependencies/configure-version-updates · https://devguide.python.org/versions/PMF prior 82/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.82/100:问题会直接中断依赖更新 PR,平台、Python、CI 与发布负责人清晰,恢复证据有经济价值;但需求是事件驱动而非高频日常使用,Dependabot 日志、CI、Renovate、手工更新和内部 runbook 都是强替代品,因此先排队验证,不占主动流量席位。Retained for a time-boxed test; PMF has not been claimed.
#0069 版在浏览器本地复核最多 20 个脱敏服务—角色行Secure MFA Proof
在线Live
74/100
中置信度medium confidence
19 / 13 / 10 / 18 / 8 / 3 / 3 按任务价值匹配one time evidence pack
尚未形成成熟行为证据launched no distribution
CISA 的权威指南证明 MFA 是重要控制;Microsoft Entra 的官方文档同时证明成熟平台已分别提供注册与实际使用报告,并能查看 MFA 挑战和方法结果。因此问题真实,但跨服务证据清单只是窄楔子,不能把安全重要性或用户手填状态当作购买与控制有效性的证据。 Evidence: https://www.cisa.gov/sites/default/files/publications/NCSAM_MFAGuide_2020.pdf · https://learn.microsoft.com/en-us/entra/identity/monitoring-health/concept-usage-insights-report · https://learn.microsoft.com/en-us/entra/identity/authentication/howto-mfa-reportingPMF prior 74/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 74/100:小企业老板、兼职 IT 与 MSP 的买家清楚,关键服务 MFA 缺口后果高;但审查频率有限,身份平台原生报告更权威,独立工具差异化弱。只有跨服务真实清单、后续复核、已验证控制变更和无关联付费共同出现,才支持继续。Retained for a time-boxed test; PMF has not been claimed.
#0097 资助方报告要求到事实交接Cause Report Brief
在线Live
76/100
中低置信度medium-low confidence
17 / 12 / 8 / 16 / 10 / 8 / 5 按任务价值匹配usage credits
尚未形成成熟行为证据launched no distribution
PMF 先验 76/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Candid describes repeated, inconsistent grant-report questions and reuse pressure; Instrumentl exposes the required period, goals, finance, signature and funder-instruction workflow. GrantHub and free templates are strong substitutes, so the narrow opportunity is a low-volume, browser-local fact handoff rather than another grant-management platform. Evidence: https://candid.org/blogs/increasing-efficiency-in-grant-reporting-nonprofit-sector/ · https://www.instrumentl.com/blog/how-to-write-a-grant-progress-report · https://smartsimple.foundant.com/products/granthub/ · https://www.instrumentl.com/templates/grant-report-template保留为限时验证;尚未声称已获得 PMF。Pre-judgment is 76/100: fragmented reporting creates real deadline and coordination pain, and a five-dollar per-report purchase can fit teams handling only a few reports. Frequency, template substitution, full-platform competition and the need for second-report reuse materially cap the prior; a polished first-session output is not PMF.
#0077 仅使用别名的借用设备事件状态机Desk Loaner Track
在线Live
74/100
中低置信度medium-low confidence
15 / 14 / 11 / 14 / 10 / 6 / 4 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Equipment checkout is a mature, proven workflow with free and paid substitutes; the narrower unresolved pain is proving that returned loaners were independently erased before reissue when endpoint actions can remain pending or fail to report. This prior is researched, not behavioral evidence for DeskLoanerTrack. Evidence: https://eqcheckout.com/ · https://www.shelf.nu/solutions/equipment-check-in · https://www.connect2software.com/ · https://learn.microsoft.com/en-us/intune/device-management/actions/delete · https://learn.microsoft.com/en-us/troubleshoot/mem/intune/device-management/troubleshoot-device-actions保留为限时验证;尚未声称已获得 PMF。PMF prior 74/100: buyer and loss-avoidance value are clear, but established ITAM and MDM suites already own checkout and device actions. The experiment earns a slot only as a deliberately narrow evidence bridge between physical return, endpoint response, independent erase verification and reissue approval.
#0076 仅使用别名的环保培训证据登记Green Training Proof
在线Live
74/100
中低置信度medium-low confidence
17 / 12 / 12 / 16 / 9 / 4 / 4 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current EPA material distinguishes generator categories and, for the relevant large-quantity context, identifies job, training-description, completion and personnel-retention records; this confirms a recurring evidence job but not universal applicability. SafetyCulture already offers completion certificates and bulk certificate workflows inside a mature training platform, showing both real demand and a strong substitute. The viable wedge is an alias-only register for micro-facilities without an LMS/EHS—not another course or compliance system. Evidence: https://www.epa.gov/hwgenerators/frequent-questions-about-hazardous-waste-generation · https://www.epa.gov/hwgenerators/hazardous-waste-generator-regulatory-summary · https://www.epa.gov/hwgenerators/hazardous-waste-generator-regulations-compendium · https://help.safetyculture.com/004126保留为限时验证;尚未声称已获得 PMF。PMF prior is 74/100, not PMF. Training changes, personnel turnover and review dates make the record job recurring and economically meaningful for a clear EHS/operations owner. Yet spreadsheets are free, LMS/EHS suites already solve more of the workflow, state and program differences create maintenance burden, and $5 supports only a narrow metadata layer. Launch is justified as an alias-only evidence-gap register with user-entered dates, never as applicability, competence or compliance certification.
#0098 源语言与目标语言分段结构和标记比较Learn Language Gap
在线Live
76/100
中低置信度medium-low confidence
17 / 11 / 11 / 14 / 11 / 7 / 5 按任务价值匹配one time structural review
尚未形成成熟行为证据launched no distribution
PMF 先验 76/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Articulate currently integrates 80+ language translation, glossaries, human validation, changed-segment updates and unified publishing, while Smartling separates tag, number, segment-completeness and placeholder checks from human LQA. These strong substitutes prove the workflow and sharply limit the wedge to a cheap structural preflight for small teams without a TMS; contextual education duties remain outside the product. Evidence: https://www.articulate.com/features/localization/ · https://www.articulate.com/360/localization/ · https://help.smartling.com/hc/en-us/articles/115003151014-Quality-Checks-Types-and-Configuration · https://help.smartling.com/hc/en-us/articles/1260806295070-LQA-Overview · https://www.ed.gov/laws-and-policy/civil-rights-laws/race-color-and-national-origin-discrimination/race-color-and-national-origin-discrimination-key-issues/equal-education-opportunities-english保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment is 76/100: multilingual release errors are real and the mechanical token job is clear, but enterprise suites already solve it, a generic diff is free, purchase frequency is episodic and three-dollar economics leave little acquisition budget. The product survives only if completed comparisons reduce real human review time and produce repeat releases or paid packs.
#0078 固定周期二元实验预注册Product Experiment Clock
在线Live
74/100
中低置信度medium-low confidence
15 / 13 / 11 / 14 / 10 / 6 / 5 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Official platform documentation validates sample planning, minimum business-cycle and peeking risks, but also demonstrates strong free calculators and platform-native statistical engines. The plausible wedge is a light preregistration and rule-change evidence layer, not sample-size calculation itself; this remains a researched prior without behavioral proof. Evidence: https://support.optimizely.com/hc/en-us/articles/4410283969165-How-long-to-run-an-experiment · https://www.optimizely.com/tools/sample-size-calculator · https://vwo.com/tools/ab-test-sample-size-calculator/ · https://help.vwo.com/hc/en-us/articles/36888587913113-Understanding-Sample-Size-Calculations · https://docs.statsig.com/experiments/advanced-setup/sequential-testing保留为限时验证;尚未声称已获得 PMF。PMF prior 74/100: teams running repeated experiments face real early-stop and after-the-fact rule-change risk, while free calculators and full platforms commoditize the arithmetic. Advance only as a narrow preregistration ledger that keeps fixed-horizon and external-engine rules separate and proves repeated end-to-end experiment use.
#0079 访谈证据分离器Product Interview Proof
在线Live
74/100
中低置信度medium-low confidence
15 / 13 / 12 / 14 / 10 / 5 / 5 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current vendor documentation confirms that interview highlights, tags, AI suggestions, source traceability, governed sensitive-data handling and evidence-quality synthesis are established platform capabilities. The remaining hypothesis is a lightweight browser-local separator for small teams, not an empty market or a superior research repository. Evidence: https://docs.dovetail.com/help/highlights · https://dovetail.com/solutions/product-research/ · https://docs.dovetail.com/help/docs · https://www.productboard.com/product-management-prompts-library/user-pain-point-synthesis/保留为限时验证;尚未声称已获得 PMF。PMF prior 74/100: mixing participant statements, behavior, observations, promises and researcher interpretation is a recurring decision risk, but mature research platforms and general AI are strong substitutes. Advance only as a privacy-first, transparent human-correction wedge; survival requires repeat interviews and changed decisions, not demo classifications.
#0075 整数分发布前压力测试Shop Promo Stack
在线Live
74/100
中置信度medium confidence
15 / 13 / 10 / 17 / 10 / 5 / 4 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
PMF 先验 74/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current Shopify documentation confirms a nontrivial application order—product discounts first, order discounts on the revised subtotal, then shipping—and documents code limits, supported channels, best-discount behavior and Plus-specific same-line combinations. The platform and discount apps already execute campaigns, so the only credible wedge is a dated, deliberately limited pre-publish contribution-margin test that makes merchant-entered costs and exclusions visible. Evidence: https://help.shopify.com/en/manual/discounts/discount-combinations · https://shopify.dev/docs/api/functions/latest/discount · https://help.shopify.com/en/manual/discounts/discount-types/buy-x-get-y · https://help.shopify.com/en/manual/discounts/discount-types/percentage-fixed-amount保留为限时验证;尚未声称已获得 PMF。PMF prior is 74/100, not PMF. A stacked promotion can create a measurable and costly margin mistake for a clear Shopify operator, and exact-intent seasonal search is reachable. But use clusters around campaigns, Shopify plus its app ecosystem are strong substitutes, custom Function and eligibility behavior cannot be emulated safely, and a $5 subscription has weak defensibility. Launch is justified only for the narrow supported arithmetic and merchant checkout reconciliation; sales prediction and campaign execution are excluded.
#0093 处理最多 5 条公开发布变化和 20 条公开文章索引Support Knowledge Expiry Clock
在线Live
76/100
中置信度medium confidence
17 / 13 / 13 / 15 / 9 / 5 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
独立检查了 Guru 验证、Intercom 内容建议和帮助中心自动更新替代品,并手工收集 GitHub、Stripe、Cloudflare、Vercel、Shopify、Notion、Linear、Slack、Figma、Jira 共 10 个公开产品的 30 条发布变化。样本覆盖新增、权限、价格、重命名、弃用和无文档影响;30/30 确定性分类复算通过。竞争表明日历复核与原生集成已拥挤,仍存在跨供应商公开发布说明到候选文章的窄切口。 Evidence: https://www.getguru.com/features/verification · https://www.intercom.com/help/en/articles/11394959-use-ai-powered-content-recommendations-to-improve-fin · https://github.blog/changelog/ · https://changelog.shopify.com/ · https://www.figma.com/release-notes/PMF prior 76/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.PMF 预判 76/100:支持知识漂移是重复且有责任人的工作,错误文档会增加工单与信任成本;但 Guru、Intercom、Document360 和帮助中心原生能力替代很强。只保留“跨产品公开发布变化→确定性复核候选”的窄定位,不承诺正确性、自动改写或发布。达到第二次真实发布周期和会员证据前,不把上线或生成队列当作 PMF。Retained for a time-boxed test; PMF has not been claimed.
#0071 版在浏览器本地复核最多 10 个脱敏资产Desk Asset Dispose
在线Live
73/100
中低置信度medium-low confidence
17 / 14 / 10 / 17 / 9 / 3 / 3 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
NIST SP 800-88 Rev.2 证明组织需要基于信息敏感度建立介质净化程序;FTC 将安全处置纳入小企业持续安全运营。Firstbase 等 ITAD 服务直接执行物流、处理、转售与证据,CertDestroy 低价销售证书,说明买家与强替代都存在;网页的独立价值只能是跨供应商缺件追踪。 Evidence: https://csrc.nist.gov/pubs/sp/800/88/r2/final · https://www.ftc.gov/business-guidance/small-businesses/cybersecurity · https://www.firstbase.com/it-asset-disposition-services · https://www.certdestroy.com/PMF prior 73/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 73/100,低于目录 77:买家与高风险任务明确,但单个小企业批次低频,真实可信度来自执行工具、独立验证和 ITAD 提供方,而不是网页状态。只有原始证据回填、缺件关闭、第二批次复用和付费同时出现,才证明独立会员价值。Retained for a time-boxed test; PMF has not been claimed.
#0128 活动标签卫士Campaign Tag Guard
在线Live
82/100
中置信度medium confidence
17 / 14 / 13 / 13 / 14 / 6 / 5 按任务价值匹配one time batch correction pack
尚未形成成熟行为证据distribution active
PMF 先验 82/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 distribution_active;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。14,800 monthly US searches for utm builder; multiple current team complaints about taxonomy drift; paid alternatives at $19–$99; exact-audit competitors and free builders keep differentiation low.82/100 预判:需求量和付费替代成立;市场拥挤使差异化仅 1/2。只验证已有混合链接的本地命名审计,不做通用 Builder。Retained for a time-boxed test; PMF has not been claimed.
#0073 版在浏览器本地准备一份脱敏升级材料Desk Vendor Escalate
在线Live
73/100
中低置信度medium-low confidence
16 / 14 / 11 / 14 / 10 / 5 / 3 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
AWS、Cisco 与 Microsoft 的官方支持文档都要求影响、范围、时间、复现、已尝试动作和诊断材料,证明证据准备是真实工作;同一文档也证明官方门户、Case Manager 和诊断访问已覆盖大量功能。独立工具只能以跨供应商最小事实、脱敏与结果回填形成窄价值。 Evidence: https://docs.aws.amazon.com/awssupport/latest/user/create-support-case-from-interaction.html · https://www.cisco.com/c/en/us/support/web/tac/technical-services-resource-guide.html · https://learn.microsoft.com/en-us/microsoft-365/admin/support-diagnostic-information-collection?view=o365-worldwide · https://www.cisco.com/c/en/us/support/docs/cloud-systems-management/application-policy-infrastructure-controller-apic/214520-guide-to-collect-tech-support-and-tac-re.htmlPMF prior 73/100 from the seven scored factors: pain, buyer clarity, frequency, economic value, distribution fit, differentiation and maintenance leverage. Evidence stage: launched_no_distribution. No behavioral PMF verdict is allowed before the app-specific clock and qualified-visit floor.投前 PMF 73/100:IT 服务台与 MSP 买家清楚,一次少来回可能节省数小时;但升级任务间歇、官方入口和 ITSM 模板替代强,供应商规则维护负担高。必须观察案例号回填、60 天复用、首次回复不补基础信息和 MSP 付费,而非只看模板下载。Retained for a time-boxed test; PMF has not been claimed.
#0092 公开表单字段采集前决策复核Trust Form Field Risk
在线Live
75/100
中低置信度medium-low confidence
16 / 12 / 10 / 15 / 10 / 7 / 5 按任务价值匹配one time decision register
尚未形成成熟行为证据launched no distribution
PMF 先验 75/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current ICO guidance states that personal data should be adequate, relevant and limited to what is necessary, while the FTC tells businesses to inventory website collection and avoid collecting unneeded sensitive information. Zoho Forms supplies strong security/lifecycle controls and PrivaScan detects fields for privacy-policy generation, but neither reviewed substitute centers a browser-local pre-collection purpose, necessity, alternative and retention-trigger decision record. Evidence: https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/data-protection-principles/a-guide-to-the-data-protection-principles/data-minimisation/ · https://www.ftc.gov/business-guidance/resources/protecting-personal-information-guide-business · https://www.zoho.com/forms/secure-forms/data-lifecycle.html · https://www.privascan.net/保留为限时验证;尚未声称已获得 PMF。PMF prior 75/100: field minimization is a real product/privacy decision before collection and the narrow workflow is safely testable. However, reviews are occasional, search demand is likely small, and secure form builders, scanners, spreadsheets and advisers are strong substitutes. The $7 pack survives only if real multi-field reviews and paid decision evidence appear—not because guidance exists.
#0094 单次事件更新边界检查Support Incident Reply Boundary
在线Live
75/100
中置信度medium confidence
18 / 13 / 8 / 17 / 11 / 4 / 4 按任务价值匹配one time preparedness pack
尚未形成成熟行为证据launched no distribution
PMF 先验 75/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Independently reviewed 12 public incident threads: six Cloudflare and six GitHub incidents spanning investigating, identified, monitoring and resolved. Manual labels tracked confirmed fact, unknown, impact scope, changed scope, cadence, mitigation, residual impact and deferred cause. Threads repeatedly added or changed scope after the first notice. PagerDuty recommends early acknowledgement, scoped impact, meaningful periodic updates, a next-update expectation and commander-confirmed recovery. Atlassian and PagerDuty already provide strong free or platform-native templates, so the narrow test is decision boundaries rather than copy generation. Evidence: https://response.pagerduty.com/during/external_communication_guidelines/ · https://support.pagerduty.com/main/docs/status-update-templates · https://support.atlassian.com/statuspage/docs/incident-template-library/ · https://www.cloudflarestatus.com/api/v2/incidents.json · https://www.githubstatus.com/api/v2/incidents.json · https://stspg.io/zj1wdthm3m6c · https://stspg.io/448g37mrq066保留为限时验证;尚未声称已获得 PMF。PMF prior 75/100. Incident communication errors have urgent economic and trust cost and a clear support/success/product buyer, but incident frequency is episodic and free Statuspage templates plus PagerDuty workflows are strong substitutes. The only defensible wedge is a vendor-neutral local hold/ready receipt that separates fact, unknown, scope, cadence and recovery evidence while forcing ETA, cause, security and remedy claims into separate human approval. This is a candidate, not proven PMF.
#0080 仅使用别名的补丁例外事件账本Secure Patch Clock
在线Live
73/100
中低置信度medium-low confidence
19 / 11 / 13 / 18 / 7 / 2 / 3 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 73/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。CISA provides an authoritative exploited-in-the-wild prioritization signal, while CIS guidance and current RMM products make automated inventory, version evidence and patch deployment the expected core workflow. A standalone manual ledger can only test the narrower delay-approval and verification-evidence gap beside those systems. Evidence: https://www.cisa.gov/known-exploited-vulnerabilities-catalog · https://www.cisecurity.org/controls/cis-controls-navigator/v8 · https://cas8.docs.cisecurity.org/en/latest/source/Controls7/ · https://www.msp360.com/rmm/patch-management/保留为限时验证;尚未声称已获得 PMF。PMF prior 73/100: unpatched exposure and approved-delay accountability have high stakes, but differentiation is only a small exception register that cannot know asset state. Advance as a weak, low-price complement; stop if incumbent RMM fields are sufficient or buyers require scanning and deployment.
#0081 仅使用别名的数据最小化决策登记Secure Data Min
在线Live
72/100
中低置信度medium-low confidence
18 / 10 / 11 / 14 / 8 / 6 / 5 按任务价值匹配membership subscription
尚未形成成熟行为证据launched no distribution
PMF 先验 72/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Official European Commission guidance separates purpose limitation, data minimisation and storage limitation; ICO asks organizations to know what they hold and why, justify periods and review data; NIST treats inventory and mapping as risk-governance inputs. These sources establish the job, not demand for this manual product. Mature discovery/privacy platforms and advisers remain strong substitutes. Evidence: https://commission.europa.eu/law/law-topic/data-protection/information-business-and-organisations/principles-gdpr_en · https://ico.org.uk/for-organisations/uk-gdpr-guidance-and-resources/data-protection-principles/a-guide-to-the-data-protection-principles/storage-limitation/ · https://www.nist.gov/privacy-framework · https://www.nist.gov/privacy-framework/using-privacy-framework-11保留为限时验证;尚未声称已获得 PMF。Pre-judgment 72/100: reducing unnecessary sensitive data has a clear loss-reduction story and named privacy/data owners, but the browser-local register cannot discover dark data or execute disposal. The viable wedge is a small, human-reviewed decision sprint for teams not ready for a platform; PMF remains unproven until repeat reviews, intent and retained subscriptions cross the preset floor.
#0095 单笔退款交接时间线Billing Refund Evidence Index
在线Live
73/100
低置信度low confidence
17 / 13 / 12 / 15 / 8 / 4 / 4 会员订阅Membership
尚未形成成熟行为证据launched no distribution
PMF 先验 73/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Current Stripe documentation confirms pending-balance, requires-action, failed, canceled, succeeded and reversal branches; customer visibility can lag about 5–10 business days, supported references can take up to 7 business days, and some failures can return within 30 days. WooCommerce distinguishes a locally recorded manual refund from actual gateway movement. Processor dashboards, receipts and full returns suites remain strong substitutes; existing CashRefundProof covers accounting reconciliation, not this support-to-processor handoff. Evidence: https://docs.stripe.com/refunds · https://docs.stripe.com/receipts · https://woocommerce.com/document/woocommerce-refunds/ · https://woocommerce.com/products/returns-for-woocommerce/ · https://10000app.rushos.com/cash-refund-proof保留为限时验证;尚未声称已获得 PMF。PMF pre-judgment 73/100. The recurring support handoff is real, but differentiation and defensibility are weak because processor and returns platforms own most evidence. Test only the cross-system promise-to-approval-to-processor-to-customer-visible timeline; do not expand into refund execution, RMA, accounting reconciliation, bank guarantees or chargeback advice.
#0082 供应商中立的 Agent 记忆合同复核Agent Memory Scope
在线Live
67/100
中低置信度medium-low confidence
17 / 9 / 7 / 11 / 10 / 8 / 5 按任务价值匹配one-time review pack
尚未形成成熟行为证据launched no distribution
PMF 先验 67/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。LangMem documents hierarchical namespaces and configurable runtime identifiers; Google Memory Bank documents scoped isolated collections, IAM and TTL; OWASP treats agentic systems, including persistent memory/context threats, as a threat-modeling problem. These sources prove that memory scope is a real design and security concern, while also showing framework-native and managed substitutes are strong. Evidence: https://langchain-ai.github.io/langmem/concepts/conceptual_guide/ · https://langchain-ai.github.io/langmem/reference/memory/ · https://docs.cloud.google.com/gemini-enterprise-agent-platform/scale/memory-bank · https://docs.cloud.google.com/gemini-enterprise-agent-platform/scale/memory-bank/setup · https://genai.owasp.org/resource/agentic-ai-threats-and-mitigations/保留为限时验证;尚未声称已获得 PMF。Pre-judgment 67/100: teams need to decide whose memory is stored, who may read or write it and how deletion is tested, but the task is often one design review and frameworks already provide the runtime primitives. The independent wedge is a vendor-neutral, content-free contract and evidence queue; willingness to pay is weak, so one-time $5 pricing is more credible than membership and PMF remains unproven.
#0096 手工标签与像素责任收据Trust Pixel Purpose Receipt
在线Live
70/100
中低置信度medium-low confidence
17 / 12 / 11 / 14 / 8 / 4 / 4 按任务价值匹配one time accountability register
尚未形成成熟行为证据launched no distribution
PMF 先验 70/100,来自痛点、买家、频率、经济价值、分发、差异化和维护杠杆七项评分。当前证据阶段为 launched_no_distribution;达到独立观察时钟和合格访问门槛前,不作行为 PMF 结论。Google Tag Manager already exposes consent-check configuration, while Usercentrics, Tracklution and AesirX make tracker discovery free or heavily substituted. FTC evidence confirms that pixel data flows create a real inventory and governance job. The narrow unresolved wedge is therefore not detection: it is a manual, human-owned receipt connecting an existing export to purpose, accountable role, public disclosure evidence, review date and a recorded change decision. Evidence: https://support.google.com/tagmanager/answer/10718549?hl=en · https://usercentrics.com/us/privacy-compliance-scanner/ · https://audit.tracklution.com/ · https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2023/03/lurking-beneath-surface-hidden-impacts-pixel-tracking · https://aesirx.io/services/privacy-scanner保留为限时验证;尚未声称已获得 PMF。PMF prior remains 70/100: buyer and risk are clear, but usage frequency, distribution and differentiation are weak. Launch only the low-cost accountability handoff after an existing scanner or tag-manager export. Do not expand into crawling, CMP enforcement, monitoring, enterprise governance or legal certification unless paid behavior independently proves that demand.
Scarce traffic allocation

同时只跑 3 个主动实验

7 / 3当前主动席位:adb-fastboot-path-diagnostic、agent-receipt、ai-context-recovery-receipt、ai-mark、broker-import-normalizer、campaign-tag-guard、prompt-inject-map
6等待账户或合作方;等待期间不计行为观察窗
RANK → SLOT完成或停止一个实验后,才按 PMF 分数补下一名
Shared account discipline

RushOS 只用一半额度

语言规则服务目标,不是机械翻译。

x · en每天最多 2 条原创 + 4 条实质回复;共享总上限 4 / 8,剩余一半保留。最小间隔 360 分钟;单 App 每周最多 2 条原创。At most 2 original posts and 4 substantive replies per day. The shared caps are 4 / 8; the other half remains reserved. Minimum gap: 360 minutes; one app gets at most 2 original posts per week.
xiaohongshu · zh-CN每天最多 1 条原创 + 2 条实质回复;共享总上限 2 / 4,剩余一半保留。最小间隔 720 分钟;单 App 每周最多 2 条原创。At most 1 original posts and 2 substantive replies per day. The shared caps are 2 / 4; the other half remains reserved. Minimum gap: 720 minutes; one app gets at most 2 original posts per week.
DAILY → WEEKLY每天采集曝光、互动、点击、合格访问、使用、意向和付款;每周只有达到预设样本的产品才可晋级或暂停。Capture reach, engagement, clicks, qualified visits, use, intent and payment daily; only apps that reach their preset sample can advance or pause in the weekly review.
Distribution before launch

逐 App 流量合同

未达到最低分发样本,只能暂停渠道,不能宣判 PMF。

access-error-message-route · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、GitHub、Web 开发社群和设计社群Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 access-error-message-route 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish an exact-intent page and W3C-grounded comparison for “form error message accessibility testing” and “WCAG error recovery checklist,” with no compliance claim. OMA: one English X teardown card and one Chinese Xiaohongshu before/after evidence story only when this app holds a weekly traffic slot; maximum cadence one original per assigned week, never quota-filling. CRO: free six-phase receipt → copied retest brief → $12 regression pack, measuring qualified visit → complete six rows → copy or checkout intent → paid. GEO: expose concise cited definitions and limitations for AI answers. Partnerships: offer the template to accessibility consultants and QA communities after two independent completions. X-Paid remains off until at least one organic checkout intent or sale; then cap a tiny exact-intent test with no retargeting.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 60 个合格访问21 days · at least 60 qualified visits

继续:Continue: 至少运行 21 天并获得 60 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 60 qualified visits, continue active distribution only if at least 12 visitors complete all six phases and either 3 run a second scenario or at least 1 genuine paid purchase occurs; keep cannot-tell and human-review boundaries intact.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At or after 120 qualified visits, pause active distribution if six-phase completion stays below 8% with zero checkout intent, or interviews show buyers only want automated scanners. Close immediately if the product cannot reliably preserve cannot-tell, avoid sensitive data, or keep certification decisions with qualified humans.

act-intake · 等待合作方Waiting for a partner

LinkedIn、创业者社群、合规伙伴内容互换Planned channels: LinkedIn.

内容:Content: 用角色/地区/领域说明为什么先做范围初筛,不卖恐慌。For act-intake, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个公司场景 + 1 次伙伴联合内容Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 每 100 合格访问 ≥1 个自愿合格线索,且伙伴愿意付费。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 合格访问 0 线索,或无合规伙伴接受线索,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

adb-fastboot-path-diagnostic · 主动实验Active experiment

计划渠道:VSO, X, Xiaohongshu, GEOPlanned channels: VSO, X, Xiaohongshu, GEO.

内容:Content: 围绕 adb-fastboot-path-diagnostic 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Teach mode → host enumeration → cable → authorization → interface path using fictional states. No flashing, unlocking, driver download link or guaranteed fix.

频次:Cadence: 按 App 独立节奏执行,额度是上限而非发满目标。D3 launch; D8 Windows fastboot path; D15 ADB unauthorized path; D24 observed funnel note. Search-first; social only when a concrete path is explainable. · 30 天 · 至少 220 个合格访问30 days · at least 220 qualified visits

继续:Continue: 至少运行 30 天并获得 220 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。At 30 days and 220 qualified visits: diagnostic completion >=15%, copy/export intent >=4%, and either one $5 purchase or two repair-handoff requests.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After 440 qualified visits: completion <8%, zero purchase and handoff intent <1%; or device-specific exceptions make the read-only tree wrong for more than 10% of reviewed cases.

agent-context-leak · 等待主动流量席位Waiting for an active traffic slot

计划渠道:X, GitHub, Dev.toPlanned channels: X, GitHub, Dev.to.

内容:Content: OSO/GEO 捕获 AI agent cross-session leakage test、tenant isolation test、agent memory isolation QA、tool context boundary 等精确意图,以 OWASP、OpenAI 和 AWS 原始文档为事实源;OMA 只有获得最多三个主动流量席位之一后,每周发布一个英文共享上下文竞态案例,并可用中文解释“同一运行时不等于同一权限边界”,不公布可利用目标;CRO 记录免费矩阵完成→批量请求→API 意向→5 美元 credits→重复消耗;X-Paid 在至少 3 个合格批量意向或 1 单真实购买前关闭。For agent-context-leak, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,若测试矩阵完成率达到 10%,且出现至少 3 个合格批量请求、API 意向达到 5% 或至少 1 个无关联 credits 购买,则继续投入框架导出与可复核用例格式。Continue only after at least 60 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到完整观察窗和 300 个合格访问后,若完成率低于 5%、没有批量意向且没有付费,或买家明确只需要实时渗透测试、运行时保护与认证而不是测试设计,则停止主动分发并关闭独立产品,不通过连接生产目标制造使用。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

agent-memory-scope · 等待主动流量席位Waiting for an active traffic slot

计划渠道:X, GitHub, Dev.toPlanned channels: X, GitHub, Dev.to.

内容:Content: 围绕 agent-memory-scope 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent English pages for agent memory namespace design, cross-user memory isolation checklist and deletion-test contract, citing framework and managed-service documentation. GEO: answer blocks that distinguish namespace strings, runtime authorization and observed deletion evidence. VSO/OMA: one redacted before/after memory-contract card on X only if the app wins one of three traffic slots; developer forums receive a technical anti-pattern, not a security claim. CRO: free 12-row review → one-time $5 pack, measuring qualified visit→contract completion→design-review use→checkout intent→paid. X-Paid stays off until organic completion and paid intent clear the gate.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days and 150 qualified visits, continue or award active traffic only if 20 teams complete real redacted contracts, eight use the artifact in an actual design review, paid intent reaches 5%, and five independent teams buy. Before both floors, zero traffic or zero sales is inconclusive.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Only after both 60 days and 150 qualified visits, stop active distribution if fewer than ten teams complete a contract, paid intent is below 2%, buyers only need framework documentation, or no team uses the artifact in a real review. Do not add hosting or enforcement to rescue a weak planning product.

agent-receipt · 主动实验Active experiment

计划渠道:X, Dev.to, LinkedInPlanned channels: X, Dev.to, LinkedIn.

内容:Content: 每次展示一张“可读收据”前后对比:工具、授权、结果、人工决定。For agent-receipt, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: X 每周 2 个英文例子;每两周 1 篇技术复盘Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 完成率 ≥10%,会员意向 ≥3% 或 1 名真实会员。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 合格访问 0 会员且意向 <1%,暂停会员与主动分发。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

agent-tool-blast · 等待主动流量席位Waiting for an active traffic slot

计划渠道:X, GitHub, Dev.toPlanned channels: X, GitHub, Dev.to.

内容:Content: 围绕 agent-tool-blast 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish an exact-intent English page for agent tool permission blast radius, approval scope and least-action design, with OWASP/Microsoft definitions and the free MCP scanner listed as a substitute. OMA: one X/Dev.to/GitHub scenario weekly showing intended action versus maximum reachable effect. CRO: expose the complete free local review before offering the $11 evidence pack. X-Paid: no spend until organic completion or paid intent appears.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue only if at least 10% complete a real tool review and there is one genuine $11 purchase, or paid intent reaches at least 3% with concrete review-pack demand.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop standalone distribution if completion is below 5%, there is no genuine purchase or qualified paid intent, users consistently prefer the free MCP scanner/platform controls, or support cost exceeds the $11 ceiling.

ai-clause · 等待合作方Waiting for a partner

小红书中文、LinkedIn、采购/法务伙伴Planned channels: Xiaohongshu, LinkedIn.

内容:Content: 用一个遗漏条款解释为什么是证据请求而非法律结论。For ai-clause, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个合同缺口案例;每月 1 次伙伴内容Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 每 100 合格访问 ≥1 条合格线索且合作方接受。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 访问 0 线索,或无可信合作方,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

ai-context-recovery-receipt · 主动实验Active experiment

计划渠道:GEO, x, VSO, X, XiaohongshuPlanned channels: GEO, x, VSO, X, Xiaohongshu.

内容:Content: 围绕 ai-context-recovery-receipt 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Show a fictional “before vague summary / after verified receipt” comparison. Never paste repository content, logs, credentials or claim automatic memory.

频次:Cadence: 按 App 独立节奏执行,额度是上限而非发满目标。D2 launch; D6 fictional handoff example; D14 native-feature comparison; D21 observed funnel note. One reusable example, adapted rather than copied across platforms. · 21 天 · 至少 160 个合格访问21 days · at least 160 qualified visits

继续:Continue: 至少运行 21 天并获得 160 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。At 21 days and 160 qualified visits: sample-to-receipt completion >=14% and either one $9 purchase or paid intent >=3%; at least five users report weekly cross-session pain.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After 320 qualified visits: completion <7%, zero purchase and intent <1%; or current AI clients provide equivalent reviewable export for most respondents.

ai-mark · 主动实验Active experiment

X、小红书、LinkedIn、创作者社群Planned channels: X, Xiaohongshu, LinkedIn.

内容:Content: 同一内容在 deepfake、AI 辅助和公共利益文本下为什么不能贴同一句标签。For ai-mark, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 法规窗口期每周 3 条,之后每周 1 条Run on an app-specific cadence; publishing caps are ceilings, not targets. · 14 天 · 至少 200 个合格访问14 days · at least 200 qualified visits

继续:Continue: 复制/生成率 ≥12%,checkout ≥2% 或一笔真实付款。Continue only after at least 14 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 合格访问仍无复制行为或付费意向,停止产品分发。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

apiauth-scope · 等待主动流量席位Waiting for an active traffic slot

计划渠道:GitHub, Dev.to, XPlanned channels: GitHub, Dev.to, X.

内容:Content: 围绕 apiauth-scope 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target exact intent around OpenAPI OAuth scopes, least-privilege API token review and operation security inheritance. OMA: publish one worked fictional contract on Dev.to/GitHub and one concise English X explanation per weekly active slot. CRO: let visitors complete the free matrix before offering the $11 workpaper; ask one optional buyer-role/use-case question. No paid acquisition before paid intent.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue only if at least 10% complete a real comparison plus either one genuine $11 purchase or paid intent from at least 3% of qualified visits; prioritize only if buyers name a repeatable API review workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge on age or zero traffic. At 300 qualified visits after the minimum window, pause if fewer than 5% complete a real comparison and there are no genuine purchases or credible buyer interviews; also pause immediately for secret-handling risk, misleading enforcement claims or a clearly dominant free substitute.

apifallback-route · 等待主动流量席位Waiting for an active traffic slot

计划渠道:GitHub, Dev.to, XPlanned channels: GitHub, Dev.to, X.

内容:Content: 围绕 apifallback-route 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: exact English intent pages for multi-provider API failover checklist, cross-provider idempotency and failback reconciliation. OMA: one fictional failure-mode teardown on Dev.to/GitHub and at most one English X post in a weekly active slot. CRO: finish the free preflight before the $19 workpaper and collect optional role/use-case intent. Partner outreach to SRE consultants only after completion evidence. No X-Paid spend before organic use or buyer interviews.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 28 天 · 至少 150 个合格访问28 days · at least 150 qualified visits

继续:Continue: 至少运行 28 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 28 days and 150 qualified visits, continue if at least 8% complete a real route preflight and there is one genuine $19 purchase or qualified paid intent from at least 3%; favor interviews that identify a repeatable architecture-review or incident-readiness workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge on launch age or zero traffic. At 300 qualified visits after the minimum window, pause standalone distribution if fewer than 4% complete a real preflight and there are no genuine purchases or credible buyer interviews; also pause for unsafe automation expectations, sensitive-input pressure or support cost beyond the one-time price.

apiretry-safety · 等待主动流量席位Waiting for an active traffic slot

计划渠道:GitHub, Dev.to, XPlanned channels: GitHub, Dev.to, X.

内容:Content: 制作 5 个‘500 不一定可重试’事故拆解、3 份 Stripe/AWS/自建 API 对比和 1 个免费重试安全检查器计划页,通过 OSO 技术词、GitHub、Dev.to 与 SRE 社区分发;以 150 个合格访问、30 次完整判定、8 个策略导出、5 次负责人访谈、2 个试点和 1 个 49 美元订阅承诺为验证线。For apiretry-safety, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 在 200 个失败场景上危险重试召回率达到 99%、安全建议精确率达到 95%,150 个合格访问产生 30 次完整判定、5 次负责人访谈、2 个团队试点且至少 1 个愿付 49 美元/月时,继续进入开发队列。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 若 200 个场景上危险重试召回率低于 99%、出现任何未标为证据不足的重复副作用建议,或 150 个合格访问不足 15 次判定、没有 2 个试点和 1 个付费承诺,则暂停并只保留教育内容。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

batch-or-now · 先验关闭Closed by prior

计划渠道:X, Dev.to, RedditPlanned channels: X, Dev.to, Reddit.

内容:Content: 展示“省 50% token 仍不该 Batch”的时限和重放案例。For batch-or-now, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个成本/时限对照案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 14 天 · 至少 150 个合格访问14 days · at least 150 qualified visits

继续:Continue: 激活 ≥10%,checkout ≥2% 或一笔真实付款。Continue only after at least 14 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 250 合格访问无付费意向,保留免费 SEO、停止 Pack。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

billing-cancel-path-audit · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、SaaS 社群、LinkedIn 与 Stripe 生态内容Planned channels: Google, Bing, LinkedIn.

内容:Content: 围绕 billing-cancel-path-audit 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for signup versus cancel friction, subscription cancellation UX audit and neutral evidence templates, citing the current FTC ANPRM/docket and Stripe docs with explicit non-legal language. OMA/VSO: at most one weekly anonymous scenario or timer demo when a traffic slot is granted; English for X/LinkedIn, a distinct Chinese version for Xiaohongshu, and no unreviewed brand shaming. GEO: expose definitions and fields. CRO: complete the free comparison before the $12 export. ASO/LBS do not fit; X-Paid remains zero until two organic qualified paid intents.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 40 个合格访问21 days · at least 40 qualified visits

继续:Continue: 至少运行 21 天并获得 40 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 40 qualified visits, continue only if at least eight users complete both path records and either two initiate the paid export or one unrelated buyer completes a purchase. Qualitative review must show that the evidence or retest pack changed a product, support or policy decision; an immature sample only extends observation and is not failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After both 21 days and 40 qualified visits, stop active distribution if full two-path completion is below 10% and there is no qualified paid intent. Also close if buyer evidence shows teams only need Stripe/Churnkey execution or professional legal review rather than a neutral evidence artifact. Zero traffic, validation activity and a young launch do not count as failure.

billing-discount-stack-guard · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、SaaS 社群、LinkedIn 与 Stripe 生态内容Planned channels: Google, Bing, LinkedIn.

内容:Content: 围绕 billing-discount-stack-guard 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: target exact English queries such as “Stripe stackable discounts order,” “subscription coupon margin calculator,” and “percent then fixed discount,” publishing the two-order 75-versus-76 example with direct Stripe citations and explicit exclusions. OMA: only with one of three weekly traffic slots, post at most one English X formula card and one independently rewritten Chinese Xiaohongshu margin story; never claim invoice accuracy. CRO: free full one-plan trace → alternate-order interaction or copied brief → $9 multi-plan pack; measure qualified visit → model → second scenario/copy/checkout intent → paid. Partner route: Stripe implementers and fractional finance operators after three completed models. X-Paid stays off until two organic pack intents or one sale, then exact-intent only with free-coupon-code negatives and a $5/day approved cap. ASO/LBS are not applicable.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 60 个合格访问21 days · at least 60 qualified visits

继续:Continue: 至少运行 21 天并获得 60 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 60 qualified visits, continue active distribution only if at least 12 visitors complete a valid model and either 3 open an alternate scenario or paid-pack intent, or at least 1 genuine $9 purchase occurs. Below the sample threshold, preserve passive search observation and do not call failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At or after 120 qualified visits, pause active distribution if valid model completion remains below 8% with zero paid-pack intent, or visitors consistently need Shopify execution plugins or free one-discount arithmetic instead. Close any platform template immediately if it cannot be reconciled with current documented test-mode behavior; never expand into account execution without new authority.

billing-refund-evidence-index · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、SaaS 社群、LinkedIn 与 Stripe 生态内容Planned channels: Google, Bing, LinkedIn.

内容:Content: 围绕 billing-refund-evidence-index 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact merchant-intent explanations for refund succeeded but not received, refund reversal versus credit, pending refund balance and when ARN becomes available, with current sources and consumer-intent negatives. GEO: expose concise state definitions and dated limitations. OMA/VSO: at most one weekly sanitized one-case teardown for SaaS support and finance operators, showing one handoff error and one decision. CRO: resolve one case free, then offer the $7 reusable queue only after a second-case need. ASO/LBS are not applicable. X-Paid remains off until organic completion, repeat use and paid intent exist; then test only merchant-qualified exact terms with a $5/day approved stop-loss.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified merchant visits, continue only if at least 10% complete a real refund timeline, at least 20% of completers return for a second case, and there is one real membership or qualified paid intent of at least 3%.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified merchant visits, stop if completion is below 5%, repeat use below 10% and there is no payment or qualified intent, or if merchants resolve every case inside their processor or returns platform. Stop immediately if value requires credentials, refund execution or bank-arrival guarantees.

books-mileage-method-compare · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、会计社群、LinkedIn 与小企业内容Planned channels: Google, Bing, LinkedIn.

内容:Content: 围绕 books-mileage-method-compare 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish English exact-intent pages for 2026 business mileage rate July change, standard mileage versus actual expenses and first-year or leased-car method constraints; add Chinese source-led explanations only where query evidence exists. OMA: scenario cards showing one evidence decision without promising savings. CRO: copied preparer checklist is the free completion hook; measure start, completion and disclosed outbound intent. Do not give an active traffic slot yet and do not run X-Paid until at least 10 organic completions and 3 monetizable outbound intents justify partner or paid-channel work.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue or improve the route only with at least 20 comparison starts, 10 completed comparisons, and either 3 qualified disclosed-partner outbound intents or 1 attributed commission; zero traffic or a young launch is never failure evidence.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the minimum-age gate, pause active distribution if 300 qualified visits produce completion below 20% or if completed comparisons create no monetizable disclosed outbound intent. Keep the free source page indexed if it earns qualified use; never infer PMF failure from validation traffic, zero traffic or insufficient age.

broker-import-normalizer · 主动实验Active experiment

计划渠道:VSO, X, Xiaohongshu, x, GEOPlanned channels: VSO, X, Xiaohongshu, x, GEO.

内容:Content: 围绕 broker-import-normalizer 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Use fictional before/after rows to show header, date, action and duplicate problems. Never show returns, recommendations, account data or tax conclusions.

频次:Cadence: 按 App 独立节奏执行,额度是上限而非发满目标。D1 launch; D4 one mapping teardown; D10 target-format comparison; D21 observed funnel note. At most one original per due platform window. · 28 天 · 至少 180 个合格访问28 days · at least 180 qualified visits

继续:Continue: 至少运行 28 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。At 28 days and 180 qualified visits: analysis completion >=12% and either one $12 purchase or paid intent >=3%; mapping-error searches must outperform broad investing traffic.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After 360 qualified visits: completion <6%, zero purchase and paid intent <1%; or most users require account sync/cost basis rather than preflight. Rotate traffic and record the reason.

cache-check · 先验关闭Closed by prior

计划渠道:X, Dev.to, RedditX、Dev.to、Reddit/LocalLLaMA

内容:Content: 用真实复用次数展示缓存从省钱变亏钱的临界点。For cache-check, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 首周 2 篇案例 + 3 次问题回复Run on an app-specific cadence; publishing caps are ceilings, not targets. · 14 天 · 至少 150 个合格访问14 days · at least 150 qualified visits

继续:Continue: 工具激活率 ≥10%,或 checkout/集成意向 ≥2%,或出现真实付款。Continue only after at least 14 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 250 合格访问后激活 <5% 且无付费意向,停止该渠道。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

campaign-tag-guard · 主动实验Active experiment

计划渠道:SEO, GEO, x, VSO, X, XiaohongshuPlanned channels: SEO, GEO, x, VSO, X, Xiaohongshu.

内容:Content: 围绕 campaign-tag-guard 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。fb/facebook and cpc/paid fragment reports; audit existing links locally rather than build another URL

频次:Cadence: 按 App 独立节奏执行,额度是上限而非发满目标。D1 launch; D3 tutorial; D7 observed-data story; D14 comparison; no more than one original per platform in a due window · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。At 21 days and 150 qualified visits: audit completion >=10% and either >=1 paid pack or paid intent >=3%.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After 300 qualified visits: completion <5%, zero payment and paid intent <1%; rotate the slot and record evidence, not an early PMF claim.

cash-audit-bundle · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-audit-bundle 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target exact English queries such as transaction evidence checklist, receipt approval audit trail and bookkeeping evidence pack, with IRS recordkeeping pages and QuickBooks/Xero audit logs cited as substitutes. OMA: publish one fictional before/after evidence-gap scenario per week on X and Dev.to; no customer figures. CRO: let users complete and copy the local queue before the $11 workpack offer. X-Paid: no ad spend until qualified organic completions or paid intent appears.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue only if at least 10% complete a real local CSV evidence queue and there is one genuine $11 purchase, or qualified paid intent reaches at least 3%.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop standalone distribution if completion is below 5%, there is no genuine purchase or qualified intent, users stay entirely inside QuickBooks/Xero/accountant workflows, or support/privacy cost exceeds the $11 ceiling.

cash-budget-variance · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-budget-variance 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO targets explain budget variance, budget variance commentary template and budget vs actual analysis with a runnable sample and explicit Xero/Fathom/LivePlan comparison. OMA publishes one anonymized “same number, different cause” case weekly; LinkedIn for finance leaders, X in English and separately written Chinese for Xiaohongshu. CRO shows the top five arithmetic drivers free, then offers the $12 full signed workpack; track qualified visit→valid mapping→human cause/evidence→export intent→paid→next-month return. Invite five bookkeepers/fractional CFOs. X-Paid remains off until 150 qualified visits and five payments.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue only with at least 70% valid mapping, 12% full analysis, 30% of completers confirming one real driver with evidence, 8% report intent and five paid packs. Only test membership if at least 25% return with a new month.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if full analysis is below 6%, nobody pays, next-month return is below 10%, at least 70% of buyers say Xero/QuickBooks/Fathom is sufficient, or users expect the system to invent causes without evidence. Never fail it before both age and sample gates.

cash-invoice-match · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-invoice-match 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target exact English queries such as match invoice payment bank statement CSV, cross-system invoice reconciliation and processor payout invoice matching; pages must distinguish this from QuickBooks/Stripe native tools. OMA: one fictional three-export mismatch walkthrough on bookkeeping communities, LinkedIn and X only when the app has an active traffic slot. CRO: let visitors match 50 rows and copy the queue before the $5/100-credit offer; capture optional source-system and monthly-row intent. Partner outreach to independent bookkeepers after real completions. No paid ads before organic qualified use.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 28 天 · 至少 150 个合格访问28 days · at least 150 qualified visits

继续:Continue: 至少运行 28 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 28 days and 150 qualified visits, continue if at least 10% complete a three-file local match and there is one genuine $5 credit purchase or qualified API/paid intent from at least 5%; prioritize if at least two users name the same recurring export pair.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge on launch age or zero traffic. At 300 qualified visits after the minimum window, pause if fewer than 5% complete a three-file match and there are no genuine purchases or credible bookkeeper interviews; also pause if most real cases require split/batch/FX logic or support/privacy cost exceeds the credit revenue ceiling.

cash-late-invoice · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-late-invoice 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: create exact-intent pages for how to prioritize overdue invoices, AR aging next actions and QuickBooks/Xero CSV review, each with a runnable sample, scoring ledger, limitations and native substitutes. GEO: publish the decision distinction between aging, dispute, promise and next action. OMA: at most one weekly anonymized queue decision or outcome through bookkeeping and small-business channels; never publish customer data. CRO: show ten useful rows before the $9 weekly pack and measure qualified visit→valid mapping→queue completion→recorded next action→paid. Start with five bookkeeper partners and template long tails, not broad social or paid search; the app must earn one of only three active traffic seats through completion and intent.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue only if CSV mapping succeeds for 70% of attempts, at least 15% complete a core queue, ten users record a real next action, paid intent reaches 8%, and five independent customers pay.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop active development and distribution if completion remains below 5%, fewer than three independent customers pay, most buyers prefer their QuickBooks/Xero native reminder workflow, or ledger/privacy constraints prevent reliable mapping and a safe recurring review.

cash-project-margin · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-project-margin 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for agency project profitability, project margin from CSV, including labor cost and cross-export contribution margin, each with a runnable example, formula and Xero/QuickBooks/Float counterexamples. GEO: publish a structured distinction among contribution, gross, operating and net profit. OMA: at most one weekly anonymized before/after cost-completeness case, rewritten for English professional channels and Chinese small-business channels. CRO: show one project and missing cost evidence free, then offer $12 multi-project monthly review. Recruit five fractional CFO/bookkeeper partners and record qualified visit→three-file map→cost confirmed→portfolio intent→paid→next-month return; no broad paid traffic until the full gate is met and the app wins an active traffic seat.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue only if valid mapping reaches 70%, at least 15% complete a project, ten users add a genuine missing cost, portfolio-report intent reaches 8%, five independent users pay, and at least 25% return the next month.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if completion is below 7%, there are no independent payments, next-month return is below 10%, 70% of buyers say Xero/QuickBooks/Float already removes the friction, or project owners cannot confirm a safe and consistent cost policy.

cash-refund-proof · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-refund-proof 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent source pages for Stripe refund reconciliation, PayPal refund fee reconciliation, match refunds to original sales and third-party fees in QuickBooks, with sample CSV and official sources. OMA: one anonymized gross/fee/net decision story weekly, English on X/LinkedIn and separately written Chinese on Xiaohongshu. CRO: free first-five exceptions then the $19 complete workpack; track valid parse→confirmed gap→report intent→paid→next-month reuse. Invite five ecommerce bookkeepers. X-Paid remains off until 150 qualified visits, precision gates and five payments are reached.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue only with at least 70% valid parses, 12% complete-chain checks, 100% exact-ID precision, at least 95% confirmed precision for fuzzy candidates, 8% report intent, five paid packs and three bookkeepers returning next month.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if completion is below 6%, there are no payments, next-month reuse is below 10%, more than 70% of buyers solve the job in one provider native report, fuzzy false positives exceed 5%, or value requires automatic posting or retained sensitive data.

cash-subscription-recon · 等待主动流量席位Waiting for an active traffic slot

搜索、会计社群、LinkedIn、小红书与伙伴Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 cash-subscription-recon 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: target “business subscription audit CSV”, “find recurring charges in bank CSV”, “software subscription price increase audit” and “subscription owner review checklist” with an exact schema, fictional worked example and explicit comparison to Ramp/Brex/Rocket Money. OMA: if granted one of three active slots, publish one anonymized English/Chinese evidence story weekly—transaction looked duplicated but belonged to two teams, annualized candidate was not savings, or a price step was usage/tax rather than vendor increase—on LinkedIn/X/Dev.to and finance-ops communities. CRO: free top-five candidates precede the $19 full review; measure valid import→owner confirmation→report intent→paid→verified action. X-Paid: $0 until organic payment and high-precision evidence; prepare exact-intent/negative terms only, with approval and CAC below gross margin before any spend.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before 21 days and 150 qualified visits. Continue only if valid parsing is at least 75%, at least 15% of qualified visitors confirm one recurring candidate with an owner, at least 10 users confirm a real duplicate/price/ownership/use issue, full-report intent is at least 8%, there are at least 5 non-affiliated $19 purchases and at least 3 verified follow-up actions. Across 20 anonymized statements, high-confidence recurring-candidate precision must be at least 95%.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At or after 300 qualified visits, stop if completion is below 7%, there is no purchase or verified action, or owner-verified candidate precision is below 90%. Stop if at least 70% of target buyers prefer Ramp/Brex/Rocket Money/native banking and will not use a redacted CSV, or if value requires bank credentials, automatic cancellation, guessing license use from transactions or presenting candidate annualized amount as savings.

cause-report-brief · 等待主动流量席位Waiting for an active traffic slot

非营利社群、LinkedIn、搜索、邮件与伙伴Planned channels: LinkedIn.

内容:Content: 围绕 cause-report-brief 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish an exact-intent page for grant report fact checklist, requirement-to-source map and nonprofit reporting handoff, then build GEO-ready answers from Candid and current reporting guides. OMA: release anonymized before/after gap cards and a transparent build decision on X in English and Xiaohongshu in Chinese only when this app wins one of three active traffic seats. CRO: let visitors finish fifteen rows without an account, expose the missing-fact queue, invite a five-dollar current-report pack, and measure a separate second-report return. X-Paid: no spend initially; after organic use-to-intent evidence, cap an exact-search test and stop on qualified-visit and paid-report economics rather than clicks.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 300 个合格访问120 days · at least 300 qualified visits

继续:Continue: 至少运行 120 天并获得 300 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 120 days and 300 qualified visits, continue only with at least 70 completed mappings, 30 paid report packs, 15 organization-level second-report purchases, a median reported preparation-time reduction of at least 50 percent, and zero fact-generation or privacy incidents.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge zero traffic or a fresh launch. At the mature 120-day and 300-qualified-visit gate, stop if fewer than 30 packs are paid or fewer than 15 organizations return for a second report, or interviews show free templates and existing grant platforms fully solve the handoff; also stop immediately for repeated requests to invent outcomes, interpret compliance, ingest sensitive records or submit reports.

cloud-backup-reach · 等待主动流量席位Waiting for an active traffic slot

GitHub、X、安全社群、LinkedIn 与搜索Planned channels: GitHub, X, LinkedIn.

内容:Content: 围绕 cloud-backup-reach 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: exact English intent for backup isolation checklist, ransomware recovery evidence, immutable vault restore test and backup RPO RTO proof. OMA: publish one fictional configured-versus-restored evidence gap on Dev.to/LinkedIn/X only while holding an active traffic slot; cite CISA and provider-native substitutes. CRO: complete and copy the free preflight before the $19 workpaper, then ask optional provider/role/drill-cadence intent. Partner route to MSP/DR consultants only after genuine completions. No paid acquisition before paid intent.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 28 天 · 至少 150 个合格访问28 days · at least 150 qualified visits

继续:Continue: 至少运行 28 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 28 days and 150 qualified visits, continue if at least 8% complete a real recovery-scope preflight and there is one genuine $19 purchase or qualified paid intent from at least 3%; prioritize if buyers name a recurring quarterly review, insurer, customer or audit evidence workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge on launch age or zero traffic. At 300 qualified visits after the minimum window, pause if fewer than 4% complete a real preflight and there are no genuine purchases or credible buyer interviews; also pause if buyers need provider execution/certification rather than an evidence workpaper or support liability exceeds the one-time price.

cloud-break-glass · 等待主动流量席位Waiting for an active traffic slot

GitHub、X、安全社群、LinkedIn 与搜索Planned channels: GitHub, X, LinkedIn.

内容:Content: 围绕 cloud-break-glass 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: exact English intent around break glass account test checklist, emergency access account quarterly review, Entra emergency access monitoring and AWS Identity Center emergency access. OMA: publish one fictional normal-SSO-outage drill gap on security/identity communities, LinkedIn and X only within an active slot. CRO: complete/copy the free preflight before the $19 drill workpaper and ask optional platform/cadence intent. Partner route to MSP/IAM consultants after genuine completions. No paid spend before buyer interviews or paid intent.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 28 天 · 至少 150 个合格访问28 days · at least 150 qualified visits

继续:Continue: 至少运行 28 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 28 days and 150 qualified visits, continue if at least 8% complete a real emergency-access preflight and there is one genuine $19 purchase or qualified paid intent from at least 3%; prioritize if buyers describe a recurring quarterly control-review or audit/customer-evidence workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge on launch age or zero traffic. At 300 qualified visits after the minimum window, pause if fewer than 4% complete a real preflight and there are no genuine purchases or credible IAM/MSP interviews; also pause for requests to store secrets, perform unapproved sign-ins or certify access without provider/drill evidence.

consent-flow · 先验关闭Closed by prior

小红书中文、产品设计社群、LinkedInPlanned channels: Xiaohongshu, LinkedIn.

内容:Content: 基于 Article 50 时点发布告知/拒绝/人工接管文案前后对比。For consent-flow, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 法规窗口每周 2 个中文场景;窗口后每周 1 个Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 生成完成率 ≥12%,1 次购买或意向 ≥3%。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 访问 0 购买且意向 <1%,并入 AIMark。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

copilot-billing-preview-retirement-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Copilot admin/FinOps/enterprise billing/平台工程社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 Copilot Billing Preview retired、August 3 migration、AI usage replacement、user budget、cost center AI credit pool、usage report export 与 billing API reconciliation 建立英文精确页和中文解释;GEO 只引用 GitHub 当前公告与官方 billing 文档。OMA:获得最多 3 个主动席位之一后,每周最多发布 1 个虚构对账故事,聚焦“headline total 相同但 scope/pool attribution 不同”。CRO:免费 3 条映射到 18 美元 100 条 evidence pack,跟踪 qualified visit→complete map→export/paid intent→paid。X-Paid:至少 3 个自然完成与 1 个合格付费意向前预算为零,只准备 exact retirement/migration 搜索词并否定 generic Copilot pricing calculator。For copilot-billing-preview-retirement-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 Copilot billing visibility 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个合格访问后,达到 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 18 美元付款,并有两个团队完成真实 Preview-to-native billing reconciliation,才继续投入。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后,完整映射率低于 5%、没有真实付款,或目标团队只用原生 AI usage/budgets/reports/API、FinOps/BI 和内部月结 runbook 且不愿为跨 owner 对账迁移证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

cost-slo · 先验关闭Closed by prior

计划渠道:X, Dev.toPlanned channels: X, Dev.to.

内容:Content: 把“月账单”改成每个成功任务的 p95 成本与超限动作。For cost-slo, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个成本 SLO 例子Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 250 个合格访问30 days · at least 250 qualified visits

继续:Continue: 会员意向 ≥3% 或 1 名真实会员。Continue only after at least 30 days and 250 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 500 访问 0 会员且意向 <1%,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

deal-comp-proof · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、销售社群、搜索与邮件内容Planned channels: LinkedIn.

内容:Content: 围绕 deal-comp-proof 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: build exact-intent pages for commission statement audit, split commission evidence, clawback calculation proof and sales commission discrepancy, avoiding generic calculator traffic. GEO: publish a structured evidence chain from plan version to trigger event, attribution, adjustment and statement. OMA: at most one weekly anonymized case where the amount looked plausible but event/version evidence was missing, adapted for LinkedIn, sales and RevOps communities. CRO: show one row detail free, then charge $9 for a full question-preparation pack; partner with independent compensation and RevOps advisors without percentage-of-dispute fees. Measure qualified visit→known row→confirmed evidence gap→authorized review→paid; do not use X-Paid until five independent purchases and the arithmetic gate are met.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue only with 20 real redacted statements, at least 95% double-checked calculation agreement, ten user-confirmed event or attribution evidence gaps, and five independent $9 purchases.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if completion is below 5%, purchases are fewer than three, real-statement review agreement is below 95%, formulas cannot be represented without material plan interpretation, or users primarily demand wage recovery, contract/legal interpretation or production CRM/payroll integration.

deal-discount-gate · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、销售社群、搜索与邮件内容Planned channels: LinkedIn.

内容:Content: 围绕 deal-discount-gate 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO targets discount impact on gross margin, profit impact of discount and give-get negotiation calculator, explicitly avoiding broad CPQ terms. OMA publishes one anonymized scenario weekly explaining why a 10% discount can surrender far more than 10% of contribution profit; English for X/LinkedIn and separately written Chinese for Xiaohongshu. CRO gives the full calculation first, then offers the $9 signed workpack; track qualified visit→complete→cost confirmed→give-get recorded→pack intent→paid. X-Paid stays off until 150 qualified visits and five payments.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 21 days and 150 qualified visits, continue only with 25 completed real scenarios, at least 20 cost-confirmed inputs, 10 written give-get conditions, paid intent of 8% or higher and five $9 paid workpacks; only then interview repeat users about a member policy/history product.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if completion is below 5%, fewer than three users pay, most users refuse to enter costs, or buyers say value requires CRM quote creation and automated approval. Do not infer failure from zero traffic or from observations before both age and sample gates.

deal-renewal-signal · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、销售社群、搜索与邮件内容Planned channels: LinkedIn.

内容:Content: 围绕 deal-renewal-signal 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For deal-renewal-signal, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Make no behavioral PMF verdict until both 30 days and 150 qualified visits. Continue only if 15 real renewal reviews achieve at least 75% CSV mapping, ten users confirm a meaningful evidence gap, and five buy the $19 pack.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After 300 qualified visits and at least 30 days, stop if completion remains below 5%, fewer than three buy, or most target buyers require real-time CRM/product integration and automated health scoring rather than an independent evidence pack. Record which substitute won; do not expand into a CS platform.

dependabot-alert-archive-transition-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Advanced Security/AppSec/GRC/平台/数据工程社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO 建立 Dependabot closed alert retention、two-year archive、UI/API missing、archive CSV admin/security manager、GHEC vs GHES 和 data residency 精确页;OMA 仅获席位后每周一个虚构历史查询断点;CRO 免费 3 条到 18 美元;三个自然完成和一个合格意向前不投放,并否定 vulnerability scanner 查询。For dependabot-alert-archive-transition-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 closed-alert history 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且至少 180 个 UTM 合格访问后,若有 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 18 美元付款,并有两个团队完成真实 archive-to-GRC 或 warehouse 历史迁移后继续。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后完成率低于 5%、没有真实付款,或团队只使用原生 CSV、SIEM/GRC 和内部 runbook 且不愿为跨系统证据交接付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

dependabot-cooldown-policy-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub/Dependabot/DevSecOps/平台工程与软件供应链社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 Dependabot default 3-day cooldown、version updates vs security updates、cooldown 1-90 days、SemVer days、include exclude precedence、GHES 3.23 建英文精确意图页;GEO 只引用 GitHub 当前资料。OMA:仅取得每周最多 3 个主动席位之一后,发布完全虚构的“普通更新延迟但安全更新立即开 PR”或“exclude 覆盖 include”策略故事。CRO:免费 3 条→策略冲突→15 美元 100 条证据包→外部付款导出;内部 QA/checkout 不计漏斗。X-Paid:3 个外部完整映射和 1 个合格意向前预算为零,之后只测 exact configuration/migration intent,否定 generic calculator。当前 queued_no_active_slot。For dependabot-cooldown-policy-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 Dependabot cooldown 策略断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线至少 30 天且达到 160 个 UTM 合格访问后才评判;继续条件是至少 25 次完整映射、8 次外部导出、5 个合格付费意向、3 笔真实 15 美元付款,并有 2 个团队用矩阵修复真实 cooldown/security-update 策略错配。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个 UTM 合格访问后,若完整映射率低于 5%、没有真实付款,或买方只用 Dependabot/Renovate/内部 policy 且不愿为跨仓库证据交接付费,则停止研发和主动流量;零流量、刚上线、内部 QA 和验证 checkout 不判 PMF。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

dependabot-python39-update-recovery-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Python/Dependabot/GitHub Actions/AppSec/平台工程社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 Dependabot Python 3.9 no update PR、unsupported Python recovery、dependency update stopped 等精确迁移意图建立英文主页面与中文解释;GEO 用官方弃用公告、官方配置文档和 Python 生命周期证据回答。OMA:获得最多 3 个主动席位之一后,每周最多发布 1 个虚构案例,讲清“job 运行但没有新 PR”与“新 PR 才是恢复证据”的决策。CRO:免费 3 条映射到 15 美元 100 条 evidence pack,跟踪 qualified visit→complete map→export/paid intent→paid。X-Paid:在至少 3 个自然完成和 1 个合格付费意向前不投放,只准备 exact-intent 搜索词,排除 generic Python upgrade/install error。For dependabot-python39-update-recovery-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 Python update recovery 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个合格访问后,达到 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 15 美元付款,并有两个真实团队用证据图恢复 Dependabot 更新 PR,才继续投入。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后,完整映射率低于 5%、没有真实付款,或目标团队只用原生 Dependabot 日志、CI、Renovate/手工更新和内部 runbook 且不愿为跨 owner 恢复证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

desk-asset-dispose · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: OSO/GEO 围绕 NIST 800-88 Rev.2 checklist、IT asset disposition chain of custody、missing destruction certificate 与 ITAD evidence tracker 建任务页,明确 Clear/Purge/Destroy 是需授权判断的外部过程;OMA 仅获主动流量席位后每月发布一例匿名“交接 20 台为何只收到 19 份序列化结果”,不展示资产标识;CRO 跟踪免费真实批次→原始证据回填→缺件关闭→第二批次→19 美元会员,并准备 MSP/认证回收商合作素材;X-Paid 在证据回填、季度复用与付费都出现前关闭。For desk-asset-dispose, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 150 天 · 至少 200 个合格访问150 days · at least 200 qualified visits

继续:Continue: 至少上线 150 天且获得 200 个合格访问后,若 12 个组织建立真实批次、8 个回填执行来源证据、5 个发现并关闭缺件或缺证、3 个无关联买家付费,且至少 2 家 MSP 在第二批次复用,则继续投入供应商协作与证据索引。Continue only after at least 150 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到完整时间与样本门槛后,若用户只生成计划或 PDF、不回填原始擦除与交接证据,重复批次少于 4 家或真实付费少于 2 家,或 ITAD 服务商门户已无摩擦覆盖全流程,则停止分发并关闭;不得以自我声明或本站报告冒充认证。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

desk-backup-user · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: 围绕 desk-backup-user 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for OneDrive Known Folder Move exclusions, endpoint backup scope audit and are user files backed up, using Microsoft scope and CISA recovery boundaries. GEO: publish a four-layer answer separating sync, version history, backup copy and verified restore. OMA: every two weeks share one anonymized decision such as a local accounting database outside Documents, never path/customer identifiers. CRO: let five-user matrices run free and show membership only after an administrator confirms a real UNKNOWN or omission; require restore-result follow-up before more distribution. Partner with MSPs, backup consultants and insurance-preparation advisors, avoid consumer cloud-storage keywords, and measure qualified visit→matrix→confirmed gap→restore sample→paid→next-quarter review for a scarce traffic seat.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 200 个合格访问120 days · at least 200 qualified visits

继续:Continue: 至少运行 120 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 120 days and 200 qualified visits, continue only if 15 organizations build real matrices, eight confirm an omission or UNKNOWN, five complete an observed restore sample, three independently pay, and two MSPs perform the next-quarter review.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the gate, stop if administrators cannot obtain trustworthy scope/exclusion metadata, fewer than five organizations complete restore tests, fewer than two pay, buyers only accept native backup-agent live verification, or filling signal gaps would require user-file upload or backup credentials.

desk-desk-memo · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: 围绕 desk-desk-memo 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For desk-desk-memo, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 200 个合格访问90 days · at least 200 qualified visits

继续:Continue: 至少运行 90 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Make no behavioral PMF verdict until both 90 days and 200 qualified visits. Continue only if 15 teams import real metadata, eight use it for four consecutive weeks, five review prior decisions, three subscribe at $15/month, and at least three recorded decisions show reproducible backlog or SLA improvement without unsupported causal claims.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After at least 90 days and 200 qualified visits, stop if fewer than five teams use it for four weeks, fewer than ten prior-decision outcomes are reviewed, fewer than two teams pay, or native ITSM scheduled reports are sufficient. Do not expand into generic BI or a connector platform to manufacture retention.

desk-leaver-proof · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: OSO:围绕 employee offboarding evidence、leaver access revocation proof、SaaS offboarding checklist 和 device return record 建立高意图页。GEO:引用 CISA 与 Microsoft,解释“禁用主账号”不代表全部 SaaS、共享链接和设备完成。OMA:发布匿名化的离职复盘故事,突出最后遗漏的一个共享账户或未归还设备及其决策。CRO:免费范围表先暴露未知项,真实事件时按人 15 美元购买证明包;给 MSP、HR 与审计顾问提供转介页。只在事件搜索和伙伴渠道测试,不投泛安全广告,不展示客户系统名称或内部细节。For desk-leaver-proof, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少 120 天、150 次合格访问后,需完成 15 个真实离职证明包、其中 8 个发现主身份之外的待关闭项、6 次付费,并由 3 名独立 MSP/审计人员抽查认为证据足以复核,才继续。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到样本后若用户只勾选而不引用外部证据、真实离职少于 8 个、付费少于 3 单,或多数客户的 Entra/1Password/IGA 已完整覆盖跨系统证明,则停止;任何敏感秘密被持久化也立即暂停。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

desk-license-recon · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: OSO 先覆盖 Microsoft 365 闲置许可证报表、Google Workspace 席位审计、SaaS 续约对账模板等任务词,页面用官方导出步骤和可复算示例建立可信度;GEO 发布“分配不等于活跃”“潜在节省不等于已实现节省”的结构化答案。OMA 每周发布一例匿名对账决策,展示哪些席位被保留、降级或取消及原因,不宣传未经确认的节省。CRO 以三供应商免费本地扫描承接搜索,结果页邀请保存月度快照并订阅,另向小型 MSP 提供可复用客户报告;不投泛关键词广告,只有自然搜索出现上传与确认行为后才测试窄词付费。For desk-license-recon, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 150 个合格访问90 days · at least 150 qualified visits

继续:Continue: 上线满九十天且累计至少一百五十个合格访问后,若至少十个组织完成真实导入、五个确认实际减少或降级席位、三个成为付费客户,并记录累计已实现节省超过五百美元,则继续投入供应商映射与 MSP 渠道。Continue only after at least 90 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到样本门槛后,若导出格式长期无法可靠映射、买方只下载表格不复访,或少于两个付费组织且没有可验证的实际席位调整,则停止主动获客;若用户核心诉求转为自动发现和自动撤权,也关闭该轻量产品而不扩成高权限平台。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

desk-loaner-track · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: 围绕 desk-loaner-track 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish the exact-intent page around “loaner laptop erase verification before reissue” and compare returned, command sent, device-reported complete and independently verified states. OMA: one anonymized state-chain card and one premature-reissue near-miss story for IT and security audiences on X/LinkedIn-style communities. CRO: free replay to paid evidence pack after a real return. No active traffic slot until portfolio ranking awards one; then run a 120-day, 200-qualified-visit test and measure replay, erase-review intent and paid organization outcomes.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 200 个合格访问120 days · at least 200 qualified visits

继续:Continue: 至少运行 120 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 120 days and 200 qualified visits, continue only if 12 organizations complete a real return-to-ready chain, 6 repeat within 30 days, 3 express willingness to pay for erase evidence or multi-client review, and 2 independently verify that the workflow prevented premature reissue.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the full sample threshold, stop or reposition if use is limited to free checkout tracking, fewer than 3 organizations review erase evidence, fewer than 2 pay, or target buyers consistently complete the job inside existing ITAM/MDM tools. Never stop because the launch is young or traffic is zero.

desk-mspboundary · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: OSO 围绕 MSP 责任矩阵、共同管理 IT RACI、GDAP 最小权限和续约范围差异建立可下载模板与任务页;GEO 发布“谁负责不等于谁有权限”“MSA、SOW、操作矩阵、技术授权”四层边界答案,引用 NIST/CISA/Microsoft。OMA 每周公开一条匿名边界决策,例如备份由谁监控、谁测试恢复、失败由谁通知,并解释证据。CRO 免费完成一个客户,双方第一次确认差异后提示多客户版本与季度复核;与 vCISO、CSP 顾问和 MSP 同行社群合作,不投泛 IT 外包词,只在共同签认、权限差异关闭和续约复用出现后扩大投入。For desk-mspboundary, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 100 个合格访问90 days · at least 100 qualified visits

继续:Continue: 上线满九十天且二百个合格访问后,若十五个 MSP 或客户完成真实矩阵、八个获得双方确认、五个关闭至少一项责任或访问差异、三个 MSP 付费,并有两个客户在季度或续约时复核,则继续扩展模板与只读映射。Continue only after at least 90 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到样本门槛后,若矩阵只被当一次性合同附件、双方确认少于五组、没有实际边界变更或付费少于两家,则停止;若用户要求法律保证或自动高权限执行,也不扩展并明确转介专业服务。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

desk-remote-risk · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: 围绕 desk-remote-risk 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish one exact-intent page for RMM access review, unattended remote access audit, MSP permission review and vendor access expiry, with CISA/NIST source summaries and an alias-only sample. GEO: publish concise definitions distinguishing inventory, permission review, log review and post-change evidence. OMA: twice weekly, share one fictional before/after queue row on X in English and one Chinese MSP permission-review explainer on Xiaohongshu only when a traffic slot is assigned; never reveal accounts or architecture. CRO: free 10-row result before checkout, then $19 monthly 100-row review pack and quarterly re-export checklist. X-Paid: no spend until at least eight real imports and three explicit paid intents; then test only exact-intent terms with a capped budget and negative keywords for remote desktop downloads, jobs and consumer support.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 200 个合格访问90 days · at least 200 qualified visits

继续:Continue: 至少运行 90 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 90 days and 200 qualified visits, continue only if 15 organizations import real redacted exports, eight confirm at least one high-risk item, five complete a permission reduction and fresh re-export, three become paying members, and two MSPs repeat the review in the next quarter.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature threshold, stop active distribution if fewer than eight organizations complete a real review, paid intent is below 2%, no permission reduction is re-exported, buyers only need a native vendor console, or maintaining export mappings costs more time than the review saves. Do not use zero traffic or launch age alone as a PMF failure.

desk-repeat-issue · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: 围绕 desk-repeat-issue 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish recurring help desk tickets, problem-management prioritization and ticket trend CSV pages with a runnable fictional sample, formulas, limitations and Atlassian/Freshservice counterexamples. GEO: provide a structured answer for why frequency is not root cause. OMA: at most one weekly anonymized merge/split decision and later recurrence result on X/LinkedIn or an MSP/ITIL community; no raw customer data. CRO: let users get five candidates before presenting $19 monthly baseline reports. Keep this app outside the three active traffic seats until an older seat underperforms; then test exact-intent search and one partner channel, recording qualified visit to import to confirmed cluster to assigned owner to paid.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 200 个合格访问90 days · at least 200 qualified visits

继续:Continue: 至少运行 90 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Only after 90 days and at least 200 qualified visits: continue if 15 organizations import real redacted tickets, eight confirm at least one cluster, five assign a permanent-fix owner, three pay, and at least two show a comparable post-fix recurrence decline of 20% or more.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the same gate, stop active development and distribution if grouping still needs heavy manual cleaning, fewer than five organizations take a repair action, fewer than two organizations pay, privacy constraints prevent a credible workflow, or buyers report that native ITSM reports already complete the same prioritization without material friction.

desk-vendor-escalate · 等待主动流量席位Waiting for an active traffic slot

IT 社群、LinkedIn、搜索、Reddit 与 MSP 伙伴Planned channels: LinkedIn, Reddit.

内容:Content: OSO 为 AWS support case best practices、Cisco TAC logs to collect、Microsoft support diagnostic access、vendor escalation template 等供应商+任务长尾建立页面,页首标出官方来源、核验日期与非官方入口;GEO 输出 who/what/when/where/why/how 事实层和最小化日志原则;OMA 仅获流量席位后每两周发布一例匿名前后对比,说明补足时间窗与移除无关日志如何减少追问,不保证响应;CRO 跟踪免费完整包→案例号回填→首次回复是否补基础信息→第二客户/60 天复用→12 美元会员;X-Paid 在重复生成和真实回填前关闭。For desk-vendor-escalate, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 200 个合格访问120 days · at least 200 qualified visits

继续:Continue: 至少上线 120 天且获得 200 个合格访问后,若 15 个团队完成真实证据包、8 个回填供应商案例号、5 个在 60 天内重复使用、3 个 MSP 真实付费,并记录至少 5 例首次回复无需补基础信息,则继续维护高频供应商包。Continue only after at least 120 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到完整时间与样本门槛后,若用户只复制免费模板、案例号回填少于 5 次、重复使用少于 4 家、真实付费少于 2 家,或官方入口已无摩擦收集全部证据,则停止分发并关闭;不得接管供应商账号或保存高敏日志来制造黏性。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

discover-aianswer-claim-index · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、X、LinkedIn、小红书与站长社群Planned channels: Google, Bing, X, LinkedIn, Xiaohongshu.

内容:Content: 围绕 discover-aianswer-claim-index 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: publish one exact-intent “brand fact source of truth” page and use Google’s own documentation to reject special-file and ranking-guarantee myths. VSO/OMA: no more than two weekly anonymized conflict→source→owner decision stories, English on X/LinkedIn and Chinese on Xiaohongshu when a traffic slot is earned. CRO: free 10 rows, prompt membership at row 11 or a second version; measure qualified visit→valid index→second version→paid intent→member. Keep it queued because only three apps may receive active traffic. X-Paid remains $0 until five real second versions and one qualified payment intent exist.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Only evaluate after at least 21 days and 150 qualified visits: continue if at least 10% complete a valid index, five unrelated users create a second version, and there is one real member or at least 3% qualified membership intent. Zero traffic or a new launch is not failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Stop or change channel if users only want an llms.txt generator, cannot provide visible sources and human owners, second-version rate stays below 5%, or 300 qualified visits produce no payment or qualified intent; do not expand into AI monitoring without recurring evidence.

discover-location-page-evidence · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、X、LinkedIn、小红书与站长社群Planned channels: Google, Bing, X, LinkedIn, Xiaohongshu.

内容:Content: 围绕 discover-location-page-evidence 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/SEO/LBS: publish an exact-intent “location facts, not city-name swaps” page, sourced doorway-policy explanation and two-location public matrix; never claim a Google uniqueness threshold. GEO: expose the field definition and original sources. OMA/VSO: one weekly English X/LinkedIn or Chinese Xiaohongshu evidence card only when a traffic slot is granted, contrasting one repeated statement with one location receipt. CRO: deliver the full two-location result before the $12 3–20 location workpack. ASO is not applicable; X-Paid stays zero until three organic completions and one qualified purchase intent.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Only after at least 30 days and 150 qualified visits, continue if free completion is at least 10%, at least 30% of completers enter evidence for three or more locations in their stated real workflow, and there is one real purchase or qualified paid intent of at least 3%. Human reviews must show the matrix prevented or corrected at least one unsupported, stale or wrongly repeated location claim.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the minimum-age gate, stop active distribution if three-plus-location workflow intent is below 5%, or after 300 qualified visits there is no purchase or qualified paid intent. Also stop if the dominant request is a duplicate-content percentage or bulk AI city-page writing rather than public fact evidence. Zero traffic, validation activity and a young launch are not failure evidence.

eval-drift · 队列中Queued

X、Dev.to、评测社群、搜索内容Planned channels: X, Dev.to.

内容:Content: 用模型/提示词变化说明旧 benchmark 为什么不再代表生产。For eval-drift, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个漂移案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 200 个合格访问30 days · at least 200 qualified visits

继续:Continue: 会员意向 ≥3% 或 1 名真实会员。Continue only after at least 30 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 400 访问 0 会员且意向 <1%,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

eval-enough · 计划中Planned

计划渠道:X, LinkedIn, Dev.toPlanned channels: X, LinkedIn, Dev.to.

内容:Content: 把“20 个案例够不够”改造成带成本和停止规则的评测案例。For eval-enough, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个 before/after rubric 案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 规划完成率 ≥10%,checkout ≥2% 或一笔团队询问。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 250 合格访问无工具激活或付费意向,停止独立付费包。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

eval-judge-calibrate · 等待主动流量席位Waiting for an active traffic slot

计划渠道:GitHub, X, Dev.toPlanned channels: GitHub, X, Dev.to.

内容:Content: 围绕 eval-judge-calibrate 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: target “LLM judge human agreement”, “Cohen kappa LLM evaluator”, “calibrate LLM-as-a-judge” and “judge confusion matrix” with an exact CSV schema, worked arithmetic and sourced limitations. OMA: if assigned an active traffic slot, publish one English disagreement teardown weekly on X/Dev.to/GitHub—why 90% agreement can still hide label imbalance, why one global score hides a weak slice, and why model/prompt changes force re-test. CRO: show the free confusion matrix before the $9 report; measure valid-pair calculation→disagreement review→report intent→paid→repeat calibration. X-Paid: prepare exact-intent and negative-keyword copy but spend $0 until organic purchase and repeat-use evidence; any test requires approval and a hard CAC ceiling below gross margin.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before 21 days and 150 qualified visits. Continue only with at least 25 valid calibration calculations, 10 disagreement-queue reviews, 5 full-report intents, 3 non-affiliated $9 purchases, and ten anonymized calibration sets where at least four lead to a changed rubric, judge prompt, threshold, slice policy or release decision. Within 60 days at least one buyer must repeat calibration after a version change.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At or after 300 qualified visits, stop if fewer than 10 valid calculations, fewer than 2 purchases, no repeat calibration or fewer than four of ten real sets change a decision. Stop sooner if at least four of five target buyers only want the capability inside their existing eval platform, or if value requires uploading evaluated content, declaring human labels to be truth, issuing an automatic pass threshold or supporting high-impact automation without domain governance and appeal.

eval-rubric · 先验关闭Closed by prior

X、Dev.to、评测社群、搜索内容Planned channels: X, Dev.to.

内容:Content: 每篇修复一条“看起来合理但无法复现”的 rubric。For eval-rubric, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个 rubric makeoverRun on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 300 个合格访问30 days · at least 300 qualified visits

继续:Continue: 检查激活 ≥12%;接入广告后 RPM >$1 或出现广告询问。Continue only after at least 30 days and 300 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 连续两个月 <1000 搜索访问,或 1000 广告展示无询问,停止广告投入。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

evidence-clock · 队列中Queued

LinkedIn、小红书、治理社群、搜索内容Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 用过期模型/供应商/评测证据展示一次审计为何会失效。For evidence-clock, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 张证据时钟;每月 1 个复核清单Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 会员意向 ≥3% 或 1 名真实会员。Continue only after at least 30 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 访问 0 会员且意向 <1%,停止会员。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

field-safety-brief · 等待主动流量席位Waiting for an active traffic slot

SEO/GEO、HVAC/电工/现场服务专业社群、伙伴内容Planned channels: SEO, GEO.

内容:Content: 每次只展示一个开工前遗漏:hazards、procedure、precautions、energy controls 或 PPE;明确工具不判断安全。For field-safety-brief, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个英文场景;没有主动流量席位时只承接自然搜索Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 合格访问达到 150 后,生成完成率 ≥10%,且 1 次真实购买或付费意向 ≥3%。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 300 合格访问且完成率 <5%、0 购买且意向 <1%,停止付费包;若高风险误用不可控则立即下线。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

ghes-support-bundle-upload-readiness-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Enterprise/平台工程/SRE/管理员和 incident response 社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO 建立 GHES support bundle August 18、minimum patch versions、ghe-support-upload rejected、cluster/geo bundle 和 unable-to-patch Support fallback 精确问题页;OMA 仅在获得三个主动流量席位之一后每周最多发布一个虚构 support continuity 断点;CRO 从免费 3 条完整地图到 18 美元证据包;三个自然完成和一个合格付费意向前 X-Paid 预算为零,只准备 exact-intent 词并否定 generic GitHub support。For ghes-support-bundle-upload-readiness-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 GHES support continuity 断点,8 月 18 日后按搜索证据降频Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个 UTM 合格访问后,若有 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 18 美元付款,并有两个团队完成真实补丁升级或 support-upload 演练后继续。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后完成率低于 5%、没有真实付款,或 8 月 18 日后搜索与使用需求快速消失,或团队只使用 CMDB、release notes、GitHub Support 与内部 runbook 且不愿为跨实例证据交接付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-actions-node24-migration-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Actions/DevOps/平台工程与 Action 维护者社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO:建立 GitHub Actions Node24 migration、Node20 removal fall 2026、forced Node24 test、macOS 13.4、ARM32 self-hosted runner 与 runs.using node24 精确页。OMA:仅获最多 3 个主动席位后,每周最多一个虚构的三方 Action 或旧 runner 失败故事。CRO:免费 3 条到 15 美元包,追踪 complete→export/intent→paid。X-Paid:三个自然完成和一个合格意向前不投放,否定 Node.js 教程。当前 queued。For github-actions-node24-migration-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 Node24 migration 边界案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个合格访问后,若有 30 次完整映射、10 次导出、6 个合格付费意向、4 笔真实 15 美元付款,并有两个团队用矩阵完成真实 Action/runner 迁移后继续。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后,若完整映射率低于 5%、没有真实付款,或团队仅使用 runner warnings、Dependabot/Renovate 与普通项目表且不愿为组合迁移证据付费,则停止产品和流量;刚上线或零流量不得提前判失败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-attestation-admission-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub/DevOps/Kubernetes/软件供应链安全社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO:建立 GitHub artifact attestation enforcement、Sigstore admission rollout、ClusterImagePolicy include/exempt conflict、namespace label 与 deny-test 精确页。OMA:只有获得最多 3 个主动流量席位之一后,每周最多一个虚构的 namespace 未启用或 include/exempt 同时匹配故事。CRO:免费 3 条到 18 美元包,追踪 complete→export/intent→paid。X-Paid:三个自然完成和一个合格意向前不投放,否定 generic signing/crypto。当前 queued。For github-attestation-admission-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 attestation admission 边界案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且至少 180 个合格访问后,若有 30 次完整映射、10 次导出、6 个合格付费意向、4 笔真实 18 美元付款,并有两个团队用矩阵安全完成真实 namespace-enforcement rollout 后继续。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后,若完整映射率低于 5%、没有真实付款,或团队仅使用 GitOps/Sigstore/Kyverno/GitHub 文档且不愿为 rollout 证据付费,则停止产品和流量;刚上线或零流量不得提前判失败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-code-quality-rollout-economics-map · 等待主动流量席位Waiting for an active traffic slot

计划渠道:Google, Bing, GitHub, xPlanned channels: Google, Bing, GitHub, x.

内容:Content: OSO/GEO:建立 GitHub Code Quality pricing、active/unique committer、AI credits、Actions cost、organization enablement、ruleset evaluate、coverage gate 与 rollout ROI 精确页并只引用官方资料。OMA:仅在获得最多 3 个主动流量席位后,每周最多发布一个虚构的成本或 outcome 决策故事。CRO:免费 3 条 cohort 到 18 美元 100 条证据包,记录 qualified visit→complete map→export/intent→paid。X-Paid:三个自然完成和一个合格意向前预算为零,只准备 exact pricing/rollout intent,否定 generic quality score、badge 与 linter。当前排队,不挤占流量席位。For github-code-quality-rollout-economics-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 Code Quality rollout economics 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且 180 个合格访问后,若有 25 次完整映射、8 次导出、5 个合格付费意向、3 笔 18 美元真实付款,并有两个团队用 cohort 证据作出真实 continue/rollback 决定后继续。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后完成率低于 5%、没有真实付款,或团队只用 GitHub 原生 dashboard/billing-impact dialog、现有 linters/quality platform 与表格且不愿为 rollout economics 证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-credential-revocation-recovery-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Enterprise/IAM/SecOps/incident response/平台工程社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 GitHub revoke SSO vs delete credentials、EMU break glass、credential type exclusions、automation impact、audit events 与 reauthorization recovery 建立英文精确页和中文解释;GEO 只引用 GitHub 当前原始资料。OMA:获得最多 3 个主动席位之一后,每周最多发布 1 个虚构案例,讲清“SSO authorization 被撤销不等于所有 credential 被删除”或“全企业动作恢复数月”。CRO:免费 3 条到 20 美元 100 条 evidence pack,跟踪 qualified visit→complete map→export/paid intent→paid。X-Paid:至少 3 个自然完成和 1 个合格付费意向前不投放,只准备 exact incident/recovery 查询并否定 token revoke API/button。For github-credential-revocation-recovery-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 credential recovery 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且至少 180 个合格访问后,达到 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 20 美元付款,并有两个企业完成真实 revocation-to-automation-recovery 演练,才继续投入。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后,完整映射率低于 5%、没有真实付款,或目标团队只用原生 UI/API、SIEM/SOAR、PAM 和内部 incident runbook 且不愿为跨 owner 恢复证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-models-retirement-migration-map · 等待主动流量席位Waiting for an active traffic slot

计划渠道:Google, Bing, GitHubPlanned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 GitHub Models retirement July 30、inference API/BYOK shutdown、brownout error、Microsoft Foundry migration、Copilot vs inference workload、provider contract checklist 建英文精确意图页;GEO 仅引用当前一手资料并标注停服后状态。OMA:仅取得每周最多 3 个主动席位之一后,发布一个虚构 brownout 暴露遗留调用或 endpoint 已换但 response/data/cost 未闭环的故事;2026-08 后若搜索/完成下降则降频。CRO:免费 3 条→迁移缺口→20 美元包→外部付款导出,内部 QA/checkout 不计。X-Paid:三次外部完成和一项合格意向前预算零;只测 exact retirement/migration,否定 model comparison/free API。当前 queued_no_active_slot。For github-models-retirement-migration-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个停服迁移断点,2026-08 后按剩余搜索证据降频Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线至少 30 天且达到 160 个 UTM 合格访问后才评判;继续条件是至少 25 次完整映射、8 次外部导出、5 个合格付费意向、3 笔真实 20 美元付款,并有 2 个团队完成真实 GitHub Models 替代切换。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个 UTM 合格访问后,若完整映射率低于 5%、没有真实付款,或停服后搜索/需求快速消失,或买方只用 provider SDK/gateway/internal runbook 且不愿为证据交接付费,则停止研发和主动流量;零流量、刚上线、内部 QA/checkout 不判 PMF。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-oidc-trust-migration-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub/DevOps/云安全社群、GitHub Discussions 和平台工程内容Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO:建立 GitHub OIDC immutable subject July 2026、repository transfer trust policy、owner/repository ID subject、aud/sub migration、cloud condition checklist 精确页。OMA:仅在获得最多 3 个主动席位之一后,每周最多发布一个虚构的仓库转移后 trust 失配/过宽故事。CRO:免费 3 条到 15 美元 100 条证据包,记录 complete→export/paid intent→paid。X-Paid:三个自然完成和一个合格付费意向前不投放,并准备 exact-intent 与 tutorial/免费 IAM generator 否定词。当前 queued,不抢占主动席位。For github-oidc-trust-migration-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 OIDC trust 迁移断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且至少 180 个合格访问后,若有 30 次完整映射、10 次导出、6 个合格付费意向、4 笔真实 15 美元付款,并有两个团队用矩阵安全完成真实仓库迁移或 opt-in 后继续。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后,若完整映射率低于 5%、没有真实付款,或目标团队仅使用 GitHub 设置、IaC 与云 IAM 且不愿为跨系统迁移证据付费,则停止产品和流量;刚上线或零流量不得提前判失败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-public-audience-access-migration-map · 等待主动流量席位Waiting for an active traffic slot

计划渠道:Google, Bing, GitHubPlanned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO 建立 stargazers API restriction、watchers 403、user subscriptions empty response、admin collaborator permission 和 aggregate/opt-in alternative 精确页;OMA 仅获主动席位后每周一个虚构 200-empty 契约断点;CRO 免费 3 条到 18 美元;三个自然完成和一个合格意向前不投放,并把 scrape/export/growth-hacking 词列为否定词。For github-public-audience-access-migration-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 audience access contract 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个 UTM 合格访问后,若有 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 18 美元付款,并有两个团队完成真实身份数据到 aggregate 或 opt-in 的迁移后继续。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后完成率低于 5%、没有真实付款,或团队只使用 aggregate counts、GitHub 原生 analytics 和内部 runbook 且不愿为跨系统证据交接付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-public-secret-exposure-response-map · 等待主动流量席位Waiting for an active traffic slot

计划渠道:Google, Bing, GitHubPlanned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 GitHub Public Monitoring attribution、verified-domain secret leak、external repo/issue/PR response、credential validity、provider revocation 与 public content removal 建立英文精确页和中文解释;GEO 只引用 GitHub 官方资料。OMA:获得最多 3 个主动席位之一后,每周最多发布 1 个虚构故事,讲清“内容已删除不等于 credential 已撤销”或“verified-domain 归因不等于 account 被管理”。CRO:免费 3 条到 20 美元 100 条 evidence pack,跟踪 qualified visit→complete map→export/paid intent→paid。X-Paid:至少 3 个自然完成和 1 个合格付费意向前预算为零,只准备 exact public-monitoring response 查询并否定 crawler/scanner。For github-public-secret-exposure-response-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 public-secret response 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且至少 180 个合格访问后,达到 25 次完整映射、8 次导出、5 个合格付费意向、3 笔真实 20 美元付款,并有两个企业完成真实 external-public-secret response,才继续投入。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后,完整映射率低于 5%、没有真实付款,或目标团队只用原生 alerts/recommendations、provider revocation、SIEM/SOAR 和内部 runbook 且不愿为跨边界证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-pwn-request-checkout-boundary · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Actions/DevSecOps/开源维护者与供应链安全社群Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 github-pwn-request-checkout-boundary 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent SEO/GEO pages for checkout v7 pull_request_target protection, fixed-SHA backport gaps, git/gh fetch coverage and two-workflow pwn-request mitigation, citing current GitHub primary sources. OMA: only after earning one of three active seats, post at most one weekly fictional before/after attack-path card to X/GitHub/Dev.to; no real repository details. CRO: free three-path result leads to the $20 release evidence pack and tracks qualified visit → complete map → export/intent → paid. X-Paid: zero until at least three natural completions and one qualified paid intent; then exact-intent search may be drafted with a small capped test and stop on query mismatch or no use. Keep queued now so it does not displace the portfolio traffic leaders.

频次:Cadence: 每周 1 个 privileged checkout 边界案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 至少运行 30 天并获得 160 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 160 qualified visits, continue only with 30 complete maps, ten exports, six qualified paid intents, four real $20 purchases and two teams that use the pack to repair a real pwn-request boundary.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 320 qualified visits, stop if completion remains below 5%, no one pays, or checkout protection, native security scanning and ordinary review solve the job without valuing the cross-path evidence handoff. Do not fail on zero traffic or freshness.

github-review-dismissal-authority-map · 等待主动流量席位Waiting for an active traffic slot

计划渠道:Google, Bing, GitHubPlanned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO:建立 restrict who can dismiss reviews、ruleset dismissal_restriction、user/team/App actor、ruleset/branch-protection layering、bypass、stale approvals、code owners 与 Rule Insights 精确页,只引用 GitHub 官方资料。OMA:仅在获得最多 3 个主动流量席位后,每周最多发布一个虚构的 layering 或 negative-test 断点故事。CRO:免费 3 条路径到 18 美元 100 条证据包,记录 qualified visit→complete map→export/intent→paid。X-Paid:三个自然完成和一个合格意向前预算为零,只准备 exact ruleset-governance intent,否定 actor picker、dismiss button 和 generic branch protection。当前排队,不挤占流量席位。For github-review-dismissal-authority-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 review authority layering 断点Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线满 45 天且 180 个合格访问后,若有 25 次完整映射、8 次导出、5 个合格付费意向、3 笔 18 美元真实付款,并有两个团队用矩阵完成真实 allow/deny review-dismissal 验证后继续。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个合格访问后完成率低于 5%、没有真实付款,或团队只用 GitHub 原生 ruleset/API/Rule Insights、policy-as-code 与内部 access review且不愿为 authority layering 证据付费时停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

github-runner-fleet-freshness-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Actions/平台工程/DevOps/SRE 社群与 Enterprise 运维内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 github-runner-fleet-freshness-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: exact-intent SEO/GEO pages for 2.329 registration versus runtime minimum, 30-day update rule, July 31/September 25 enforcement, critical paused jobs, audit-log incomplete inventory, update service/proxy and cached-image rebuild, citing current GitHub sources. OMA: only after earning one of three active seats, publish at most one weekly fictional freshness failure card to X/GitHub/Dev.to with no fleet identity. CRO: free three-group map leads to the $18 evidence pack and tracks qualified visit → complete map → export/intent → paid. X-Paid: zero until three natural completions and one qualified paid intent, then only exact enforcement queries with generic installation negatives and a strict small cap. Keep queued now.

频次:Cadence: 每周 1 个 runner freshness 断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 至少运行 30 天并获得 160 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 160 qualified visits, continue only with 30 complete maps, ten exports, six qualified paid intents, four real $18 purchases and two teams that use the pack to avoid a real runner job-queue disruption.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 320 qualified visits, stop if completion remains below 5%, no one pays, or teams use GitHub annotations/APIs, ARC, CMDB and image automation without valuing the cross-system evidence handoff. Zero traffic or freshness is not PMF failure.

github-untrusted-cache-boundary-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub Actions/DevSecOps/软件供应链安全社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO/GEO:建立 GitHub Actions read-only cache、untrusted trigger cache poisoning、pull_request_target save warning、fork workflow_run cascade 与 trusted cache-save workflow 精确页。OMA:仅获最多 3 个主动席位后,每周最多一个虚构的 untrusted writer→trusted restorer 或静默 save 回归故事。CRO:免费 3 条到 18 美元包,追踪 complete→export/intent→paid。X-Paid:三个自然完成和一个合格意向前不投放,否定 generic cache performance。当前 queued。For github-untrusted-cache-boundary-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 cache trust 边界案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 上线满 30 天且至少 160 个合格访问后,若有 30 次完整映射、10 次导出、6 个合格付费意向、4 笔真实 18 美元付款,并有两个团队用矩阵安全修复真实 cache trust 边界后继续。Continue only after at least 30 days and 160 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 320 个合格访问后,若完整映射率低于 5%、没有真实付款,或团队仅使用 GitHub warning、安全扫描与普通 workflow review 且不愿为跨 workflow 证据付费,则停止产品和流量;刚上线或零流量不得提前判失败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

gov-board-packet · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、小红书、专业社群与搜索Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 围绕 gov-board-packet 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For gov-board-packet, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 40 个合格访问30 days · at least 40 qualified visits

继续:Continue: 至少运行 30 天并获得 40 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Make no behavioral PMF verdict until both 30 days and 40 qualified visits. Continue only if that sample produces at least eight valid previews, three export intents, one paid $9 report, and then one same-team repeat within 60 days; otherwise adjust the message or channel without calling zero traffic product failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After at least 30 days and 100 qualified visits, stop if there is no paid report, if average buyer-reported manual rewrite exceeds 30 minutes, or if the target buyer consistently requires sensitive GRC integration before the draft is useful. Record the evidence and reason; do not restart a closed candidate.

green-project-roi · 等待主动流量席位Waiting for an active traffic slot

行业搜索、LinkedIn、环境社群与专业伙伴Planned channels: LinkedIn.

内容:Content: 围绕 green-project-roi 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: target “energy efficiency ROI calculator”, “compare HVAC retrofit quotes”, “simple payback vs NPV” and “cost of delaying energy project” with public formulas, a fictional comparison and DOE/ENERGY STAR sources. OMA: if given one of three active slots, publish one LinkedIn/industry-community case weekly where optimistic/base/conservative assumptions change rank, or where quote scope/M&V evidence changes the decision; avoid generic green claims. CRO: free two-project comparison precedes the $12 memo; measure real quote entry→comparison completion→approval-package intent→paid→documented decision. Partner material: prepare a transparent referral one-pager for contractors, energy auditors and utility programs, but no undisclosed commission. X-Paid: $0 until organic paid evidence; prepare exact-intent/negative keywords and require approval plus CAC below gross margin.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before 60 days and 150 qualified visits. Continue only if at least 30 users enter real redacted quotes, comparison completion is at least 25%, approval-memo intent is at least 6%, there are at least 3 non-affiliated $12 payments and at least 10 users report a documented approve/revise/defer decision traceable to the comparison or evidence gaps.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the gate, stop if real quote entry is below 10%, paid intent below 2%, there is no non-affiliated payment, or most users only need the free ENERGY STAR/ORNL calculator. Stop if contractors will not provide comparable assumptions, or if users expect us to estimate engineering performance, guarantee savings, determine incentive/tax eligibility or replace a qualified M&V plan.

green-training-proof · 等待主动流量席位Waiting for an active traffic slot

行业搜索、LinkedIn、环境社群与专业伙伴Planned channels: LinkedIn.

内容:Content: 围绕 green-training-proof 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for hazardous waste training record template, RCRA training evidence log and environmental training matrix using EPA sources, explicitly separating LQG examples from SQG/VSQG and state duties. GEO: provide a dated answer block explaining that a completion date alone omits job, plan, evidence and retention context, without legal conclusions. OMA: only after earning one of three traffic slots, publish one alias-only paper-sign-in-to-evidence-index before/after story every two weeks on LinkedIn/X and a Chinese operations checklist on Xiaohongshu; no compliance claims. CRO: deliver the full ten-record review first and present membership only when a second review date or personnel change exists. Prioritize transparent referrals from waste-training providers, environmental consultants and small manufacturing associations. Do not buy broad social traffic; test paid exact-intent search only after 20 organizations enter real alias-only records and at least 15% activate a user-defined reminder.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 120 个合格访问60 days · at least 120 qualified visits

继续:Continue: 至少运行 60 天并获得 120 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days and 120 qualified visits, continue only if 25 organizations enter real alias-only personnel records, 30-day organization retention reaches 25%, concrete paid intent reaches 5%, and three organizations maintain paid subscriptions for two consecutive months. Privacy-pattern regression accuracy must remain 100%, and no customer may rely on the tool to infer applicability or competence.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature sample, stop the standalone product and retain only the free template if real-record entry is below 10%, most qualified buyers already use an LMS/EHS, reminder activation is below 15%, or fewer than two subscriptions exist. Pause immediately for personal-data leakage or compliance-certification confusion. Zero traffic and young age are not failure evidence.

green-utility-recon · 等待主动流量席位Waiting for an active traffic slot

行业搜索、LinkedIn、环境社群与专业伙伴Planned channels: LinkedIn.

内容:Content: 围绕 green-utility-recon 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish monthly utility reconciliation, missing bill tracker and Portfolio Manager meter gap pages with exact interval examples and strong-substitute links. GEO: explain why bill periods are not natural months and how start/stop, replacement, corrected and consolidated bills change interpretation. OMA: publish an anonymized 12-site interval visualization only when a real case exists, never account or meter identifiers. CRO: let three facilities scan free, then offer $9 for thirty-site owner/evidence reports; target energy consultants, property accountants and multi-location operations rather than generic sustainability audiences. Measure qualified visit→valid CSV→real issue confirmed→owner assigned→paid→second-month retained; no X-Paid until the threshold and active-traffic-seat rules are met.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days and 150 qualified visits, continue only if upload success is at least 35%, twenty real accounts complete scans, paid intent reaches 6%, and three independent subscriptions remain active for two months.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the sample gate, stop or narrow to a free CSV checker if real upload is below 15%, buyers already find Portfolio Manager/accounting tools sufficient, fewer than three subscriptions persist, or PDF/vendor format demands create unsustainable manual maintenance or require credentials.

human-gate · 队列中Queued

X 英文、小红书中文、Dev.to、产品安全社群Planned channels: X, Xiaohongshu, Dev.to.

内容:Content: 每条内容比较“允许 / 单次批准 / 双人批准”的边界。For human-gate, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个动作矩阵例子Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 完成率 ≥10%,决策包意向 ≥3% 或 1 次购买。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 合格访问 0 购买且意向 <1%,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

incident-brief · 队列中Queued

LinkedIn、Dev.to、AI 治理社群、小红书Planned channels: LinkedIn, Dev.to, Xiaohongshu.

内容:Content: 把“猜根因”改成先冻结事实、时间线与影响。For incident-brief, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个匿名化事件模板Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 120 个合格访问30 days · at least 120 qualified visits

继续:Continue: 完成率 ≥8%,1 次购买或 2 个专业询问。Continue only after at least 30 days and 120 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 240 访问 0 意向,停止独立产品并并入治理包。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

label-copy · 等待账户Waiting for accounts

X、小红书、LinkedIn、创作者/营销社群Planned channels: X, Xiaohongshu, LinkedIn.

内容:Content: 一份内容生成三种错误标签,再给可复制中英文版本。For label-copy, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 法规窗口每周 3 条,之后每周 1 条Run on an app-specific cadence; publishing caps are ceilings, not targets. · 14 天 · 至少 200 个合格访问14 days · at least 200 qualified visits

继续:Continue: 复制率 ≥15% 且捐助点击 ≥1%,或一笔捐助。Continue only after at least 14 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 500 次有效生成无捐助,保留免费工具、停止维护投入。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

learn-language-gap · 等待主动流量席位Waiting for an active traffic slot

搜索、教师社群、YouTube、小红书与伙伴Planned channels: YouTube, Xiaohongshu.

内容:Content: 围绕 learn-language-gap 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: publish source-led exact-intent pages for translation placeholder consistency, number consistency, tag consistency and why structural checks do not prove translation quality, citing current Articulate and Smartling workflows. OMA: when this app earns one of three active traffic seats, publish one synthetic source-to-target omission card weekly—English on X and Chinese on Xiaohongshu—with the decision and boundary visible. CRO: let visitors complete fifteen rows without an account, then offer a three-dollar release workpack and track second-release reuse separately. X-Paid stays at zero until at least two organic purchases or five qualified paid intents; if unlocked, use exact-match localization-QA terms with strict negative terms for translation services, legal advice and free generic diff.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Only after at least 30 days and 150 qualified visits, continue if at least 18 real comparisons are completed, there are two paid packs or five qualified paid intents, manually measured critical-token miss rate is below 5 percent, and qualified reviewers report at least 25 percent lower review time. Preserve second-release reuse as a separate proof.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Do not judge a young or zero-traffic launch. At 300 qualified visits after the minimum-age gate, stop active distribution if there is no purchase, completion stays below 6 percent, users only need a free generic diff or machine translation, human review time does not fall by 25 percent, or demand concentrates on semantic, legal, accessibility or education judgments the product cannot safely make.

local-or-api · 先验关闭Closed by prior

搜索内容、Hugging Face 社群、XPlanned channels: X.

内容:Content: 按隐私/运维/负载拆真实决策,不做通用排行榜。For local-or-api, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每两周 1 个场景比较Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 300 个合格访问30 days · at least 300 qualified visits

继续:Continue: 官方出站率 ≥8%,正式联盟后出现佣金。Continue only after at least 30 days and 300 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 600 访问出站 <3%,或无合规合作计划,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

mail-oneclick-unsubscribe · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、邮件投递社群、ESP/营销运营社群和开发者内容Planned channels: Google, Bing.

内容:Content: OSO/SEO/GEO 围绕 RFC 8058 checker、List-Unsubscribe-Post missing 和 Gmail one-click unsubscribe test 建立英文精确问题页与来源驱动 FAQ;OMA 获得流量席位后每周最多发布一个虚构的‘字段通过但端点未实测’案例,X 英文优先、小红书使用独立中文版本;CRO 先完成一个免费模板,再展示 7 美元 50 模板包;ASO/LBS 不适用。三次自然完整使用和一个合格付费意向前 X-Paid 为零,之后也只允许经批准的 exact-intent 小额止损测试。For mail-oneclick-unsubscribe, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个头部失败案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 上线满 30 天且至少 150 个合格访问后,若有 30 次完整检查、10 次多模板导出、5 个合格付费意向、3 笔真实 7 美元付款,并有两名发送负责人用证据修复真实模板,则继续主动分发。Continue only after at least 30 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 300 个合格访问后完整检查率仍低于 5%、没有真实付款,或团队只使用 ESP 原生验证且不愿为静态与运行证据分离付费时停止;任何泄露真实订阅者、生产令牌或擅自调用端点的情况立即暂停。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

margin-gate · 先验关闭Closed by prior

无新增分发Planned channels: high-intent search, relevant communities and proof-led social content.

内容:Content: 复盘:为什么 $1 静态利润报告先停。For margin-gate, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 停止Run on an app-specific cadence; publishing caps are ceilings, not targets. · 0 天 · 至少 0 个合格访问0 days · at least 0 qualified visits

继续:Continue: 只有出现新的高意向入口或产品转为持续成本监控才重开。Continue only after at least 0 days and 0 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 结构性分数 <45,立即停止买量、发帖和维护。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

mcp-permit · 等待重叠问题处理Waiting for overlap review

计划渠道:GitHub, Reddit, Dev.toPlanned channels: GitHub, Reddit, Dev.to.

内容:Content: 用单个权限组合解释为何远程、宽文件、任意网络和密钥访问会叠加。For mcp-permit, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 张威胁卡 + 3 次配置点评Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 200 个合格访问21 days · at least 200 qualified visits

继续:Continue: 检查完成率 ≥10%,2 个赞助询问或 1 个真实赞助。Continue only after at least 21 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 500 定向访问无赞助意向,停止赞助销售。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

mobile-developer-verification-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Android 与替代商店开发社群、GitHub、移动发布和账户运营内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-developer-verification-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: exact-intent pages for Android developer verification, package registration status, Play vs non-Play and signing-key ownership using dated primary sources. OMA: at most one fictional responsibility-break story weekly only after a traffic seat. CRO: free three-row map to $12 pack. X-Paid stays off until three natural completions and one qualified intent.

频次:Cadence: 每周 1 个验证责任断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete maps, 10 exports, six qualified paid intents, four real $12 purchases and two teams using the matrix in a real package handoff.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion stays below 5%, no one pays, or target teams rely on Console, Studio and APIs without valuing a separate cross-account package/key evidence handoff.

mobile-play-billing-state-route · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Android 与后端开发社群、GitHub、订阅增长和收入工程内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-play-billing-state-route 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Queued without an active traffic seat. OSO: exact-intent pages for Google Play purchase lifecycle checklist, RTDN entitlement state, three-day acknowledgement refund and pending purchase handling. GEO: source-dated factual blocks citing Google and distinguishing RTDN signals from latest API state. OMA: after earning a slot, at most one fictional payment-state versus entitlement failure story weekly for English developer channels. CRO: free five-row map to $15 100-row pack and measure qualified visit → valid route → export intent → paid. X-Paid stays zero before three organic completions and one qualified paid intent; exclude generic integration and RevenueCat searches.

频次:Cadence: 每周 1 个购买状态断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete state routes, ten exports, six qualified paid intents, four real $15 purchases and two teams that use the route to repair evidence in a real Android purchase release.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion remains below 5%, nobody pays, or teams consistently finish with Play Billing and Developer APIs, RevenueCat and backend tests without valuing a separate cross-endpoint handoff. Pause distribution if current Google behavior cannot be safely reflected.

mobile-play-data-safety-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Android 与隐私工程社群、GitHub、移动发布和 SDK 治理内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-play-data-safety-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: exact-intent pages for Google Play Data safety SDK checklist, collected versus shared, SDK disclosure responsibility and declaration updates with dated primary sources. OMA: after earning a traffic seat, at most one fictional SDK-change responsibility story weekly. CRO: free three-row map to the $15 pack. X-Paid stays off until three natural completions and one qualified paid intent.

频次:Cadence: 每周 1 个 SDK 数据责任断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete maps, 10 exports, six qualified paid intents, four real $15 purchases and two teams repairing a real release declaration or responsibility gap.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion stays below 5%, no one pays, or target teams use Play Console, SDK Index and existing scanners without valuing a separate cross-component version evidence handoff.

mobile-privacy-manifest-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、iOS 开发社群、GitHub、移动隐私与发布工程内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-privacy-manifest-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Queued without an active traffic seat. OSO: publish an exact-intent Apple privacy manifest component checklist and answer pages for missing SDK manifests, required-reason API ownership and release reconciliation. OMA: one anonymized before/after component responsibility card for English X or relevant developer communities, with no claim of Apple approval. CRO: free three-row map to $12 100-row editable pack; measure qualified visit → valid map → export intent → paid. GEO: maintain source-dated factual answer blocks citing Apple. X-Paid remains off until organic completion and paid-intent evidence; no broad social cadence while only three portfolio apps may receive active distribution.

频次:Cadence: 每周 1 个 SDK 责任断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue only with 25 complete component maps, eight exports, five qualified paid intents, three real $12 purchases and two mobile release owners who use the queue to repair evidence in a real release.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if completion remains below 5%, nobody pays, or teams consistently finish the job with Xcode, App Store Connect and SDK-native reports without valuing a separate cross-component handoff. Requirement changes that cannot be corrected safely also pause distribution.

mobile-review-access-handoff · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Android/iOS 发布社群、QA 与移动工程内容Planned channels: Google, Bing.

内容:Content: 围绕 mobile-review-access-handoff 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: exact-intent pages for Play app access sign-in details, reusable review login, external IdP, subscription wall, MFA/OTP and location review access with dated official sources. OMA: after earning a seat, at most one fictional access-responsibility failure story weekly with no credentials. CRO: free three-path map to $12 pack. X-Paid stays off until three natural completions and one qualified intent.

频次:Cadence: 每周 1 个审核访问责任断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete maps, 10 exports, six qualified paid intents, four real $12 purchases and two teams using the matrix to repair a real review-access failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion stays below 5%, no one pays, or teams rely on store fields, secret managers and QA checklists without valuing a separate credential-free version handoff.

mobile-subscription-event-route · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、iOS 与后端开发社群、GitHub、订阅增长和收入工程内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-subscription-event-route 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。Queued without an active traffic seat. OSO: publish exact-intent pages for App Store Server Notifications V2 checklist, billing retry entitlement route, idempotency and notification-history recovery. GEO: maintain source-dated answer blocks citing Apple and distinguish historical delivery from current state. OMA: after earning a slot, publish at most one fictional event-to-entitlement failure story weekly for English developer channels. CRO: free five-row map to $15 100-row pack; measure qualified visit → valid route → export intent → paid. X-Paid remains zero until three organic completions and one qualified paid intent; exclude generic StoreKit and RevenueCat implementation searches.

频次:Cadence: 每周 1 个订阅事件断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete event routes, ten exports, six qualified paid intents, four real $15 purchases and two teams that use the route to repair evidence in a real subscription-backend release.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion remains below 5%, nobody pays, or teams consistently complete the job with App Store Server Library, RevenueCat and backend tests without valuing a separate cross-team event handoff. Pause distribution if Apple behavior changes cannot be safely reflected.

mobile-target-api-deadline-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、Android 与发布工程社群、GitHub、移动维护和产品运营内容Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 mobile-target-api-deadline-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: exact-intent pages for Google Play target API 2026 deadline, API 36 updates, existing App availability, form-factor differences and extension checklist with dated primary sources. OMA: after earning a seat, at most one fictional missed-form-factor deadline story weekly. CRO: free three-row map to $12 pack. X-Paid stays off until three natural completions and one qualified intent.

频次:Cadence: 每周 1 个 target API 责任断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 至少运行 45 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 45 days and 180 qualified visits, continue only with 30 complete maps, 10 exports, six qualified paid intents, four real $12 purchases and two teams completing a real multi-App or form-factor migration handoff.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 360 qualified visits, stop if completion stays below 5%, no one pays, or teams use Play Console and existing project management without valuing a separate portfolio date and test responsibility handoff.

model-license · 队列中Queued

X、GitHub、Hugging Face 社群、小红书Planned channels: X, GitHub, Xiaohongshu.

内容:Content: 不解读法律结论,只展示模型卡/许可证/权重来源的证据缺口。For model-license, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个证据缺口例子Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 工具完成率 ≥10%,1 次真实购买或意向 ≥3%。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 访问 0 购买且意向 <1%,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

npm-publish-authority-migration-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、npm/Node/GitHub Actions/开源维护者与软件供应链社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 npm bypass2fa 2026 August、direct publish January 2027、trusted publishing OIDC、staged publishing 2FA、granular token retirement 建立英文精确意图页,并用同一来源事实构造 GEO 可引用问答。OMA:仅在每周组合评审获得 3 个主动席位之一后,发布一个完全虚构的双窗口权限迁移案例和一张 decision matrix,不泄露架构或用户数据。CRO:免费 3 条即时结果→展示剩余 authority gap→18 美元 100 条证据包→付款后导出,不以内部 checkout 或自测计收入。X-Paid:在至少 3 个外部完整映射和 1 个合格付费意向前预算为零;达门槛后才测试 exact-match 搜索,不做宽泛安全广告。当前排入 queued_no_active_slot,不挤占已有最多 3 个主动流量席位。For npm-publish-authority-migration-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 npm publish authority 断点案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线至少 45 天且达到 180 个 UTM 合格访问后才评判;继续条件是至少 30 次完整映射、10 次外部导出、6 个合格付费意向、4 笔真实 18 美元付款,并有 2 个 maintainer 团队用该交接完成真实发布 authority 迁移。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个 UTM 合格访问后,若完整映射率低于 5%、没有真实付款,或买方只用 npm 原生 trust/stage 加内部 runbook 且不愿为跨包证据交接付费,则停止研发和主动流量;零流量、刚上线、内部 QA 与内部 checkout 都不能判 PMF 成败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

npm-readonly-recovery-boundary-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、npm/Node/GitHub Actions/开源维护者与软件供应链 incident 社群Planned channels: Google, Bing, GitHub.

内容:Content: OSO:围绕 npm high-impact account 72-hour read-only、email change、2FA recovery code、publish blocked、token/visibility/org-team paused、install remains available 和 post-window review 建立英文精确意图页;GEO 只把一手来源拆成可引用问题答案。OMA:仅在每周组合评审取得最多 3 个主动席位之一后,发布一个完全虚构的“依赖者仍可安装但维护者不能紧急发布”连续性案例,不展示架构或真实身份。CRO:免费 3 条→显示暂停动作和连续性缺口→15 美元 100 条证据包→外部付款后导出;内部 QA/checkout 不计漏斗。X-Paid:至少 3 个外部完整映射和 1 个合格意向前预算为零,达到门槛后才测试 exact-match recovery/read-only intent,并否定 generic npm outage/status。当前 queued_no_active_slot。For npm-readonly-recovery-boundary-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 npm read-only recovery 边界案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 45 天 · 至少 180 个合格访问45 days · at least 180 qualified visits

继续:Continue: 上线至少 45 天且达到 180 个 UTM 合格访问后才评判;继续条件是至少 25 次完整映射、8 次外部导出、5 个合格付费意向、3 笔真实 15 美元付款,并有 2 个 maintainer 团队在真实 72 小时只读连续性或窗口后恢复复核中使用证据包。Continue only after at least 45 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 360 个 UTM 合格访问后,若完整映射率低于 5%、没有真实付款,或买方只用 npm alert/Support、trusted publishing 和内部 runbook 且不愿为恢复证据交接付费,则停止研发和主动流量;零流量、刚上线、内部 QA 与验证 checkout 均不能判 PMF 成败。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

npm-v12-install-boundary-map · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、npm/Node.js/GitHub/DevSecOps/软件供应链社群Planned channels: Google, Bing, GitHub.

内容:Content: 围绕 npm-v12-install-boundary-map 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: exact-intent SEO/GEO pages for npm v12 allowScripts, implicit node-gyp blocked, git/file/link prepare, allow-git/remote none, approve/deny, project versus global policy and strict/dangerous overrides using current primary sources. OMA: only after earning one of three active seats, at most one weekly fictional migration failure card on X/GitHub/Dev.to. CRO: free three-path map to $15 evidence pack; track qualified visit → complete map → export/intent → paid. X-Paid: zero until three natural completions and one qualified paid intent, then exact migration terms with generic npm error negatives and strict cap. Keep queued now.

频次:Cadence: 每周 1 个 npm v12 install boundary 案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 160 个合格访问30 days · at least 160 qualified visits

继续:Continue: 至少运行 30 天并获得 160 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 160 qualified visits, continue only with 30 complete maps, ten exports, six qualified paid intents, four real $15 purchases and two teams repairing a real npm v12 install or CI migration.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 320 qualified visits, stop if completion remains below 5%, no one pays, or teams use npm warnings/approve-scripts, SCA and CI without valuing the cross-context evidence handoff. Do not fail on freshness or zero traffic.

private-route · 等待联盟环境Waiting for affiliate access

LinkedIn、Dev.to、隐私工程/采购社群Planned channels: LinkedIn, Dev.to.

内容:Content: 按“敏感数据 + Files + 搜索”画出三条必须核对的数据路径。For private-route, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个架构场景 + 5 次专业回复Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 官方出站 CTR ≥8%,正式联盟后出现可归因佣金。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 合格访问出站 CTR <3%,停止联盟申请。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

product-evidence-age · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: 围绕 product-evidence-age 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For product-evidence-age, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 160 个合格访问120 days · at least 160 qualified visits

继续:Continue: 至少运行 120 天并获得 160 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Make no behavioral PMF verdict until both 120 days and 160 qualified visits. Continue only if 12 teams review real evidence, eight complete a human status review, five change an evidence scope or linked decision, three buy the $9 pack, and at least three teams return for another review in a later month.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After at least 120 days and 160 qualified visits, stop if users only perform a one-time age check, fewer than five teams change evidence status, fewer than three return, or fewer than two buy. Do not add a full research repository merely to manufacture retention; record the mature evidence and closure reason.

product-experiment-clock · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: 围绕 product-experiment-clock 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target exact English questions about how long to run an A/B test, fixed-horizon peeking and preregistration rule changes, while showing formulas and current provider sources. GEO: publish the concise distinction between sample clock, business-cycle clock and engine-specific stopping rules. OMA: twice weekly, show one anonymized “rule written before vs changed after seeing results” decision story on product/data communities, LinkedIn and X only if awarded one of three active traffic slots. CRO: free calculation and first freeze before a $9 repeat audit pack; partner with experimentation/CRO consultants only after real readouts. No paid ads until real start-to-readout completion and paid intent.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 300 个合格访问120 days · at least 300 qualified visits

继续:Continue: 至少运行 120 天并获得 300 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 120 days and 300 qualified visits, continue only if 30 teams create real preregistrations, 15 return to record readouts, 8 use it across three experiments, 3 pay, and at least 5 documented cases show the frozen rule prevented early stopping or exposed a rule change.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature sample threshold, stop if visitors use only the free calculator, fewer than 10 readouts are recorded, fewer than 5 teams repeat across three experiments, or fewer than 2 pay. Do not stop for young age or zero traffic, and do not rescue it by inventing unsupported statistical guarantees.

product-feature-cost · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: 围绕 product-feature-cost 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For product-feature-cost, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 200 个合格访问120 days · at least 200 qualified visits

继续:Continue: 至少运行 120 天并获得 200 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not make a behavioral verdict before both 120 days and 200 qualified visits. Continue only if 15 teams estimate real features, eight return with matching actual costs, five explicitly change scope, sequence or priority, and three unrelated workspaces pay $15 per month.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature sample, stop standalone distribution if users only make one estimate, fewer than eight return actuals, fewer than three teams report a decision change, fewer than two unrelated workspaces pay, or most buyers prefer Jira, Linear or engineering-insight tools; do not expand into project management to hide weak demand.

product-interview-proof · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: 围绕 product-interview-proof 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for user interview evidence, facts versus interpretations and commitment bias, with definitions and source-backed examples. GEO: expose a compact evidence hierarchy that AI answers can cite. OMA: twice weekly, post one fully fictional/redacted before-and-after note where a future promise is downgraded after reviewing past behavior, on product communities, LinkedIn and X only when awarded an active slot. CRO: first analysis without email, then offer repeat cross-interview evidence packs; target PMs, founders and designers without research repositories. No paid ads before real second-interview return and paid intent.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 180 个合格访问90 days · at least 180 qualified visits

继续:Continue: 至少运行 90 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 90 days and 180 qualified visits, continue only if 20 users process real redacted interviews, 10 return with a second interview within 30 days, 6 teams correct suggestions, 4 confirm a changed requirement judgment, and at least 3 maintain paid subscriptions.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature threshold, stop if inputs are mainly fictional, fewer than 5 users return, buyers only want generic summaries without source traceability, or fewer than 2 pay. Never stop for young age or zero traffic, and do not expand into recording, recruitment or a full research repository to rescue weak demand.

product-launch-gate · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: 围绕 product-launch-gate 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For product-launch-gate, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 180 个合格访问90 days · at least 180 qualified visits

继续:Continue: 至少运行 90 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before both 90 days and 180 qualified visits. Continue only if 20 teams use real pending releases, ten invite cross-functional review, six delay or change a launch because of a discovered gate gap, and four unrelated buyers purchase the $12 pack.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature sample, stop standalone distribution if users only download a checklist, fewer than four teams report an actual go/no-go change, fewer than two unrelated buyers pay, or most users require native feature flags and monitoring. Do not expand into deployment or observability to manufacture retention.

product-metric-contract · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: OSO 捕获 metric definition template、conversion numerator denominator、ratio metric specification、proxy metric contract 等精确意图;OMA 仅在获得三个主动流量席位之一后发布英文“same metric name, opposite decision”和中文“同名指标也可能得出相反决策”的案例拆解;CRO 以免费合同→双角色批准→下游导入→版本回访→会员付费为漏斗;未验证 LTV 前关闭 X-Paid。For product-metric-contract, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 200 个合格访问120 days · at least 200 qualified visits

继续:Continue: 在 120 天且至少 200 个合格访问后,若获得至少 15 个真实团队使用、8 次产品与数据双角色批准、5 次下游导入、5 个真实口径冲突导致计算或决策改变,并至少 3 个付费会员,则继续投入版本与导出能力。Continue only after at least 120 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到观察窗和样本门槛后,若用户只把它当模板、下游导入少于 5 次、版本回访少于 4 次、付费少于 2 个,或没有任何团队提出语义层扩展需求,则停止主动分发并关闭独立产品,保留可迁移的合同模块。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

product-sunset-plan · 等待主动流量席位Waiting for an active traffic slot

LinkedIn、产品社群、X、搜索与公开构建Planned channels: LinkedIn, X.

内容:Content: OSO 建立 feature deprecation plan、product sunset checklist、service shutdown communication、data migration before feature removal 与中文“功能下线计划”等精确意图页;GEO 发布可引用的 T-90/T-60/T-30/T-7/T+0 阶段模型并链接 LaunchDarkly 和 Atlassian 原始资料;OMA 仅获主动流量席位后每两周拆解一个公开弃用公告,说明通知、迁移、数据和支持的充分或缺失证据;CRO 免费风险矩阵不设邮箱墙,检测到真实日期与跨职能审阅需求时展示 49 美元工作包;X-Paid 在至少两次自然付费前关闭。For product-sunset-plan, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 180 个合格访问120 days · at least 180 qualified visits

继续:Continue: 至少上线 120 天且获得 180 个合格访问后,若 15 个团队为真实待下线功能完成清单、8 个生成跨职能时间线、5 个邀请审阅并报告发现可行动遗漏,且至少 3 个无关联团队购买 49 美元工作包,则继续投入模板和审阅体验。Continue only after at least 120 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到完整观察窗和样本门槛后,若多数用户只下载免费清单、没有真实停止日期、发现可行动遗漏的团队少于 5 个或真实付费少于 2 次,则停止主动分发并关闭;不得扩展为通知发送、客户数据、合同审查、数据操作或法律合规平台。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

prompt-inject-map · 主动实验Active experiment

计划渠道:X, GitHub, Dev.toPlanned channels: X, GitHub, Dev.to.

内容:Content: 画出不可信内容到高权限写操作的一条真实攻击路径。For prompt-inject-map, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: X 每周 2 张路径卡;每两周 1 篇拆解Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 200 个合格访问21 days · at least 200 qualified visits

继续:Continue: 检查完成率 ≥10%,2 个赞助询问或 1 个真实赞助。Continue only after at least 21 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 500 定向访问无赞助意向,暂停销售;内容互动低于组合中位数一半则让出席位。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

red-team-plan · 等待合作方Waiting for a partner

X、GitHub、安全社群、专业伙伴Planned channels: X, GitHub.

内容:Content: 每次补一个常被遗漏的工具、记忆或恢复测试。For red-team-plan, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个测试范围拆解Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 每 100 访问 ≥1 合格线索且伙伴接受。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 访问 0 线索或无伙伴,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

retention-lens · 计划中Planned

LinkedIn、Dev.to、Hacker News 评论、隐私/安全社群Planned channels: LinkedIn, Dev.to.

内容:Content: 用“ZDR 不等于无状态、无文件、无第三方留存”拆解具体架构。For retention-lens, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个供应商场景 + 5 次专业回复Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 工具激活 ≥8%,至少 1 个企业线索或真实付款。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 合格访问无激活且 0 线索,转为内容页并停止静态包。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

retry-budget · 先验关闭Closed by prior

无新增分发Planned channels: high-intent search, relevant communities and proof-led social content.

内容:Content: 复盘:为什么工程常识不应包装成低价独立产品。For retry-budget, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 停止Run on an app-specific cadence; publishing caps are ceilings, not targets. · 0 天 · 至少 0 个合格访问0 days · at least 0 qualified visits

继续:Continue: 只有与真实遥测和框架配置集成后重开。Continue only after at least 0 days and 0 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 结构性分数 <45,立即停止分发和付费入口。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

schema-meter · 队列中Queued

GitHub、Dev.to、Reddit/MCP、代理框架讨论区Planned channels: GitHub, Dev.to, Reddit.

内容:Content: 公开匿名 schema 前后对比:描述、枚举和可选字段占多少 token。For schema-meter, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个 schema 拆解 + 1 个 API 示例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: API credits 点击 ≥3% 或首个付费 credit 包。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 次免费分析无 API 意向,停止付费 API。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

secure-asset-list · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-asset-list 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。For secure-asset-list, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not make a verdict before both 60 days and 150 qualified visits. Continue if at least 30 organizations enter 10 or more real assets, free completion reaches 25%, paid intent reaches 5%, and three unrelated customers pay.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the mature sample, stop active distribution if real-data entry is below 12%, paid intent is below 2%, most users choose Snipe-IT or MDM instead, or the dominant request is automatic scanning without a credible recurring budget.

secure-data-min · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-data-min 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: one exact-intent English source page around data minimization review checklist, purpose register and storage-limitation review, with official citations. VSO/OMA: one anonymized before/after decision card on X when a traffic slot opens; no fear-based compliance claim. GEO: concise answer blocks distinguishing inventory, necessity decision and verified disposal. CRO: free 15-row review to $5 repeat pack, tracking qualified visit → completed register → pricing/checkout intent → paid. X-Paid stays off until organic completion and intent clear the gate; no legal-services keywords or broad paid traffic.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 180 个合格访问90 days · at least 180 qualified visits

继续:Continue: 至少运行 90 天并获得 180 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 90 days and 180 qualified visits, continue or grant a traffic slot only if 25 organizations review real redacted areas, at least 20% return for a second review, paid intent reaches 6%, and three subscriptions remain active for two consecutive months. Before both time and sample floors, zero activity is inconclusive.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Only after both 90 days and 180 qualified visits, stop active distribution if fewer than 12 organizations finish a register, paid intent is below 2%, users primarily demand automatic discovery/deletion, or policy maintenance costs more than the review saves. Never close solely because launch traffic is zero or young.

secure-incident-plan · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-incident-plan 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for small business incident response plan, ransomware pocket card, cyber-insurance IRP and tabletop checklist, linking stable CISA/FTC guidance and explaining the line between plan, live technical response and legal notification. GEO: provide a structured answer emphasizing real authority, offline access and exercises over document length. OMA: tell one anonymized tabletop gap story, such as an unreachable MSP after-hours reference, adapted by platform and never exposing contacts. CRO: leave official resources and gaps free, charge $12 only for customized printable cards. Test MSP, insurance broker, privacy-counsel and association referrals before broad social; measure qualified visit→ten controls→real reference configuration→tabletop→paid, and award a scarce traffic seat only on verified configuration/intent.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 150 个合格访问90 days · at least 150 qualified visits

继续:Continue: 至少运行 90 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 90 days and 150 qualified visits, continue only if 30 independent organizations configure real contact references, 15 conduct a tabletop, paid intent reaches 7%, and five non-affiliated customers complete payment.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the sample gate, stop or convert to disclosed expert referral if real-reference configuration is below 15%, paid intent is below 3%, buyers only download empty templates, the product duplicates IncidentBrief behavior, or users expect forensic commands, notification decisions or other unsafe automation.

secure-insurance-prep · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: OSO:围绕 cyber insurance application prep、renewal evidence checklist、MFA scope answer 建事实型指南;GEO:引用 FTC、承保方公开说明和申请示例,强调准确限定而非“答得好看”;OMA:公开“邮件 MFA 是全员还是仅管理员”如何改变答案的故事;CRO:免费缺口预览后售一次性证据包,并为经纪/MSP提供有披露的联合模板。主动流量优先续保季和 10–100 人专业服务企业。For secure-insurance-prep, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 上线至少 90 天并获得 150 次合格访问后,若 30 个企业建立真实事实包、至少 15 个完成经纪或 MSP 复核、准备时间中位数可下降且有 5 笔非关联付费,则继续。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到样本后若用户只寻找免费表单、真实证据索引率低于 20%、付费意图低于 3%,或经纪/承保方认为输出增加误述风险,则停止或转为人工顾问辅助。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

secure-mfaproof · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: OSO 捕获 MFA audit checklist、MFA evidence template、registered vs enforced MFA、small business MFA review 等精确意图,并以 CISA 与身份平台官方报告为事实源;OMA 只有在获得最多三个主动流量席位之一后,按两周一篇发布“registered is not enforced or used”的英文证据拆解与中文“小企业 MFA 三种状态别混为一谈”,不展示真实账号;CRO 跟踪免费真实台账→证据引用→高关键缺口→控制变更→9 美元购买→季度回访;未证明付款与控制变更前关闭 X-Paid。For secure-mfaproof, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 120 天 · 至少 180 个合格访问120 days · at least 180 qualified visits

继续:Continue: 在 120 天且至少 180 个合格访问后,若至少 15 家组织创建真实跨服务台账、8 家在后续周期回来复核、5 位业主或 MSP 因发现的缺口完成并验证控制变更,且至少 3 个无关联买家购买 9 美元证据包,则继续投入导入与复核体验。Continue only after at least 120 days and 180 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到完整观察窗且至少 360 个合格访问后,若真实台账少于 5 个、后续复核少于 3 次、已验证控制变更少于 2 次或无关联付费少于 2 单,则停止主动分发并关闭独立产品;不得通过高权限 OAuth、账号扫描或自动修复制造使用量。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

secure-passkey-recovery · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 精确搜索、GitHub、Dev.to、身份工程与产品安全社群Planned channels: Google, Bing, GitHub, Dev.to.

内容:Content: OSO/SEO/GEO 围绕 passkey account recovery checklist、passkey rollout test plan、delete all passkeys recovery 建立英文精确问题页和来源驱动 FAQ;OMA 在获得流量席位后每周最多发布一个“更强登录却有更弱恢复”的匿名失败旅程,X 英文优先、小红书使用独立中文版本;CRO 先交付完整三场景结果,再展示 $9 五阶段包;ASO/LBS 不适用。X-Paid 只保留 exact-intent 与 IdP 品牌否定词草案,在出现自然完整使用和真实付款意向前预算为零。For secure-passkey-recovery, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个失败旅程Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 上线满 30 天且至少 150 个合格访问后,若有 30 次完整场景检查、10 次导出、5 个团队级付费意向、3 笔真实 9 美元付款,并有 2 个团队在真实发布评审中复用,则继续投入。Continue only after at least 30 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 300 个合格访问后完整检查率仍低于 5%、没有真实付款,或访谈显示团队只使用现有 IdP 与 FIDO 文档且不愿为跨阶段证据包付费时停止;刚上线、零流量和验证 checkout 不计失败或需求。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

secure-patch-clock · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-patch-clock 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target exact English intents for patch exception register, KEV remediation evidence and failed patch follow-up, explicitly contrasting the ledger with RMM deployment. GEO: publish the boundary between KEV prioritization, asset applicability, remediation report and independent verification. OMA: twice weekly, show one fictional “delay is not a moved date” event chain for MSP and security audiences only if awarded an active traffic slot. CRO: free ten-exception replay, then $5 repeat evidence review at the first due-date handoff; use MSP and audit-preparation partners rather than competing for patch-management keywords. No paid acquisition before real exception reuse and payment intent.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 120 个合格访问60 days · at least 120 qualified visits

继续:Continue: 至少运行 60 天并获得 120 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days and 120 qualified visits, continue only if 20 organizations register real exceptions, at least 10 close on time, paid intent reaches 6%, and 3 maintain paid subscriptions for two consecutive months.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature threshold, stop if users mainly require automatic scanning or deployment, fewer than 10 real exceptions are registered, paid intent is below 2%, or target RMM products already provide sufficient approval evidence. Never stop because the launch is young or traffic is zero.

secure-payment-fraud · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: OSO:围绕 vendor bank change verification、payment change callback procedure 和 BEC payment fraud 建立场景页、免费政策模板与常见失败示例。GEO:引用 FBI 最新 BEC 定义与损失,并清楚区分流程留痕、账户所有权验证和反欺诈保证。OMA:每周发布一则去标识的“邮件看似正常但联系方式来源错误”决策故事,不展示银行信息。CRO:免费政策模板引导用户在下一次真实变更时以 9 美元开案;向簿记员提供首案演示和转介码。先测试会计/簿记伙伴与 exact-intent 搜索,达到使用信号前不买广告。For secure-payment-fraud, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 100 个合格访问21 days · at least 100 qualified visits

继续:Continue: 上线至少 90 天并获得 150 次合格访问后,需有 20 次政策生成、至少 12 个真实变更案件完成、5 个付费案件,且至少 3 名簿记员或老板在第二次事件中复用,才继续投入提醒和团队功能。Continue only after at least 21 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到样本后,如只有模板下载而真实开案少于 5 个、付费少于 3 单、用户更需要银行账户所有权验证,或现有会计软件已无摩擦覆盖该流程,则停止;出现完整银行数据被误填也要立即暂停整改。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

secure-recovery-test · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-recovery-test 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for backup recovery test plan, restore exercise checklist, RPO RTO validation and disaster recovery test evidence, using CISA/NIST sources and clear “planning versus execution” language. GEO: publish a compact definition set separating backup presence, restore job completion, business validation, observed RPO/RTO and recoverability. OMA: at most twice monthly when assigned a traffic slot, share one fictional dependency or stop/rollback mistake on X in English and a distinct Chinese recovery-drill checklist on Xiaohongshu; no infrastructure details. CRO: deliver the eight-step plan without an email wall, then offer a $12 one-time editable pack; invite voluntary practitioner feedback rather than selling leads. X-Paid: no spend before five completed real plans and two paid intents; if earned, test exact-intent terms only with exclusions for backup software downloads and consumer data recovery.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 90 天 · 至少 120 个合格访问90 days · at least 120 qualified visits

继续:Continue: 至少运行 90 天并获得 120 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 90 days and 120 qualified visits, continue only if 15 authorized real exercise plans are completed, four packs are purchased, two MSP or IT practitioners confirm the steps are safe and useful, and at least five users return with observed exercise evidence.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At the mature gate, stop active distribution if fewer than eight plans are completed, no authorized exercise is actually run, fewer than two packs sell, practitioners reject the safety boundary, or users need vendor-native execution rather than a planning artifact. Zero traffic and launch age alone are never stop evidence.

secure-vendor-risk · 等待主动流量席位Waiting for an active traffic slot

搜索、LinkedIn、本地商家社群与可信 IT 伙伴Planned channels: LinkedIn.

内容:Content: 围绕 secure-vendor-risk 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish the exact-intent page for “vendor security assessment checklist for small business”, “vendor risk tiering” and “supplier due diligence questions”, with NIST/CISA definitions and clear comparison against free templates/full TPRM. GEO: maintain concise sourced definitions and a fictional worked example. OMA: under an active traffic slot, publish one redacted “the SOC badge was not the relationship scope” teardown weekly on LinkedIn/X, plus a practical GitHub/Dev.to checklist; never name a real vendor. CRO: free triage shows concrete missing evidence before the $5 one-credit workpaper; measure triage→evidence-request→paid, not email capture. X-Paid: prepare exact-intent and negative-keyword tests, but spend $0 until organic use and at least one non-affiliated purchase; then require approval and a hard CAC ceiling below gross margin.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 150 个合格访问60 days · at least 150 qualified visits

继续:Continue: 至少运行 60 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before 60 days and 150 qualified visits. Continue only if at least 30 organizations complete a real vendor triage, at least 40% use the output to request evidence, paid intent is at least 7%, and there are at least 5 non-affiliated $5 purchases; qualitative reviews must confirm the workpaper changes or documents a real procurement decision rather than merely downloading a generic questionnaire.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the gate, stop if real completion is below 15%, paid intent below 3%, no non-affiliated purchases, or most users only want the free CISA/NIST template or an enterprise platform. Stop immediately if support demand requires RushOS to certify vendors, verify private reports, interpret law/contracts, scan third parties without authorization or auto-approve a supplier.

shadow-ai · 等待合作方Waiting for a partner

小红书中文、LinkedIn、安全治理社群Planned channels: Xiaohongshu, LinkedIn.

内容:Content: 匿名化场景:个人账号 + 敏感数据 + 写权限如何排序。For shadow-ai, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个盘点场景 + 每月伙伴内容Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 每 100 访问 ≥1 合格线索且伙伴接受。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 访问 0 线索或无伙伴,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

shop-bundle-margin · 等待主动流量席位Waiting for an active traffic slot

小红书、TikTok、商家社群、搜索与平台生态Planned channels: Xiaohongshu, TikTok.

内容:Content: 围绕 shop-bundle-margin 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish the exact-intent page around bundle contribution margin, bundle break-even price and why higher AOV can reduce contribution, with Shopify primary sources and calculator examples. GEO: create a compact answer block distinguishing AOV, gross profit and contribution margin. OMA: once a traffic slot is earned, post one anonymized before/after cost-stack card weekly on X and one Chinese merchant decision story on Xiaohongshu, never promising profit. CRO: the free result exposes one decision gap, then offers a reusable review after calculation rather than before value. No paid traffic until organic visitors complete at least 30 valid calculations; then test only a tightly matched merchant keyword cohort. Continue or stop by qualified merchant behavior, not impressions.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 300 个合格访问60 days · at least 300 qualified visits

继续:Continue: 至少运行 60 天并获得 300 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days, 300 qualified merchant visits and 120 complete calculations, continue only if cent-level regression truth remains 100%, at least 35 merchants report a price, component or publish decision change, 20 save or repeat a review, 12 return after a cost change, and there are either eight real payers or three verified buyers willing to pay at least $19 for a stronger workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Immediately stop distribution on any unexplained one-cent rule error. At the full sample threshold, stop or merge the wedge if fewer than 15 merchants change a decision, 30-day repeat use is below 15%, or there is zero real payment. Never treat zero traffic or a newly launched age as failure, and never claim that AOV proves profit.

shop-promo-stack · 等待主动流量席位Waiting for an active traffic slot

小红书、TikTok、商家社群、搜索与平台生态Planned channels: Xiaohongshu, TikTok.

内容:Content: 围绕 shop-promo-stack 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: target Shopify discount combinations calculator, stacked discount margin, BFCM promotion profit and promotion margin test, with a dated rule matrix and exact worked examples. GEO: publish a concise answer distinguishing valid combination, customer discount and merchant contribution. OMA: only after earning one of three active traffic slots, publish one anonymized one-error/one-decision card per week during the six weeks before major sale periods on X, and a Chinese merchant margin story on Xiaohongshu; never claim more orders. CRO: give the complete five-line calculation before asking for payment, then offer repeat reviews when a second campaign or cost change appears. Partner with ecommerce accountants, fractional CFOs and promotion-app agencies through a co-branded checklist. Do not buy generic ecommerce traffic; only consider exact-intent seasonal paid search after 30 organic complete simulations and a confirmed decision-change rate of at least 20%.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 60 天 · 至少 300 个合格访问60 days · at least 300 qualified visits

继续:Continue: 至少运行 60 天并获得 300 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 60 days, 300 qualified merchant visits and 100 complete simulations, continue only if supported arithmetic regression accuracy remains 100%, at least 30 merchants confirm a real margin risk, 20 save or review a second promotion, 14-day repeat use reaches 25%, and there are either eight real $5 payers or three verified buyers willing to pay at least $19 for a deeper workflow.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。Pause distribution if any current official rule change cannot be corrected within seven days or any supported truth case differs by one cent. At the sample gate, stop or merge if fewer than ten of 100 simulations change a confirmed decision, or 300 qualified visits produce no repeat use and no real payment. Zero traffic and young age are never failure evidence, and unsupported sales-growth predictions cannot be used to keep the product alive.

stack-rfp · 队列中Queued

LinkedIn、Indie Hackers、采购/AI 创业社群Planned channels: LinkedIn.

内容:Content: 把“看 demo”改成五条可淘汰供应商的证据请求。For stack-rfp, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个 RFP 反例 + 1 个模板片段Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: RFP 完成率 ≥8%,出现 1 个真实收录或 2 个供应商询问。Continue only after at least 30 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 买方访问无工具使用,或 500 访问无收录意向,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

sunset-watch · 等待关键词调整Waiting for keyword revision

计划渠道:X, Dev.to, GitHubPlanned channels: X, Dev.to, GitHub.

内容:Content: 每次官方退役更新做“受影响谁、剩几天、替代怎么测”的日期卡。For sunset-watch, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 次总表;距截止 30 天内每周 3 次Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 订阅点击 ≥3% 或出现 1 名真实会员。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 300 合格访问且无订阅点击,停止会员维护。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

support-incident-reply-boundary · 等待主动流量席位Waiting for an active traffic slot

SaaS 社群、CX 社群、LinkedIn、X 与搜索Planned channels: X, LinkedIn.

内容:Content: 围绕 support-incident-reply-boundary 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO: publish exact-intent pages for first outage update, known-versus-unknown incident communication, unsupported recovery ETA and resolution evidence, each citing current primary guidance and the 12-thread decision ledger. GEO: expose concise answer blocks and public source links without claiming a universal cadence. VSO/OMA: at most one weekly public thread teardown, English-first on X and separately written Chinese on Xiaohongshu, showing one changed claim and the resulting human decision; never reuse live confidential incidents. CRO: free local check → copyable receipt → $12 tabletop/preparedness pack; ask whether a real role used it in a tabletop or incident. X-Paid remains off until at least 15 organic completions, three qualified purchase intents and one real purchase. No active traffic slot is assigned at launch.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Review only after at least 21 days and 150 qualified visits. Continue if there are at least 20 completed checks, five preparedness-plan uses, three qualified $12 purchase intents, one real purchase, and one buyer confirms use in a tabletop or real incident. Before that, zero traffic or age is not a negative PMF verdict.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop the standalone paid product if completion is below 10%, no real purchase exists, buyers only copy the free Atlassian templates, or requested value depends on live status integrations, automated legal/security claims, AI certainty or autonomous publishing. Adjust the channel instead of closing when qualified use exists but acquisition source is weak.

support-knowledge-expiry-clock · 等待主动流量席位Waiting for an active traffic slot

SaaS 社群、CX 社群、LinkedIn、X 与搜索Planned channels: X, LinkedIn.

内容:Content: OSO:围绕 “release notes documentation update checklist”“support docs stale after product release” 建立英文意图页与 10 产品公开样本方法页;GEO:发布可引用的 30 项标签方法、反例和边界。OMA:LinkedIn/X 面向 Support Ops、CX 与 SaaS 产品团队,每周最多 1 个真实发布→文章决策案例,不占用当前 3 个主动流量席位前不主动分发。CRO:免费映射后邀请用户登记下一次 release 日期并导出一次队列,第二次 release 才展示 $9/月价值;不做付费投放,直到出现 3 个合格会员意图。For support-knowledge-expiry-clock, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少上线 21 天且达到 150 个合格访问后,再要求 20 次有效公开映射、5 个团队在 45 天内处理第二个 release、3 个合格会员意图,并至少 1 个会员跨第二次发布或第二个账期仍保留,才继续研发或分配主动流量席位。Continue only after at least 21 days and 150 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 达到 300 个合格访问后完成率仍低于 10%,或没有任何第二次 release 复用/会员意图,则停止主动分发;若用户普遍选择 Guru/Intercom/原生 KB 提醒,或有用结果必须依赖私有工单、自动语义判断或直接发布,也关闭产品并公开说明。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

token-unit-cost · 等待主动流量席位Waiting for an active traffic slot

计划渠道:X, Dev.to, GitHubPlanned channels: X, Dev.to, GitHub.

内容:Content: 围绕 token-unit-cost 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: publish exact-intent pages for “cost per successful AI task”, “LLM unit economics” and “AI agent retry cost”, with a sourced definition, CSV schema and a worked arithmetic example. OMA: publish three English redacted counterexamples on X/Dev.to/GitHub—per-call price fell while cost per success rose, retry waste, and unknown outcomes distorting the denominator—at most two scenarios weekly only if TokenUnitCost earns an active traffic slot. CRO: free 1,000-row calculation precedes the $10 credit offer; measure schema use→valid import→success definition→report intent→paid. X-Paid: prepare exact-intent/negative-keyword copy but spend $0 until an organic non-affiliated purchase and a repeat processing need; any test needs approval and a CAC ceiling below gross margin.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 150 个合格访问21 days · at least 150 qualified visits

继续:Continue: 至少运行 21 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Do not judge before 21 days and 150 qualified visits. Continue only if there are at least 18 valid imports, 10 explicit success definitions, 7 report exports or intents, 2 non-affiliated $10 payments, and eight anonymized ledgers with at least seven within 2% of a human/provider-bill reconciliation; at least four ledgers must reveal a high-cost task that changes a routing, retry, feature or pricing decision.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。After the gate, stop if fewer than eight valid imports and no payment; if fewer than two of five target buyers can provide a reliable task outcome; if reconciliation error exceeds 5% in two or more of eight ledgers; or if buyers overwhelmingly prefer the same metric inside their current observability platform. Stop immediately if producing value would require raw prompts/outputs, production provider credentials, invented success labels or treating optional task value as proven revenue.

tool-allowlist · 队列中Queued

计划渠道:GitHub, Dev.to, XPlanned channels: GitHub, Dev.to, X.

内容:Content: 展示一个过宽工具策略如何收紧为域名/动作/token/deny。For tool-allowlist, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 2 个策略 diffRun on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 200 个合格访问21 days · at least 200 qualified visits

继续:Continue: ≥5% API 意向或 1 个 credit 包购买。Continue only after at least 21 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 400 访问 0 API 意向,停止并并入 MCPPermit。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

trace-gap · 队列中Queued

计划渠道:GitHub, Dev.to, XPlanned channels: GitHub, Dev.to, X.

内容:Content: 用一条无法复盘的 trace 展示缺失字段和隐私边界。For trace-gap, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个 trace makeover + 1 个 API 示例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 21 天 · 至少 200 个合格访问21 days · at least 200 qualified visits

继续:Continue: ≥5% 用户请求 API 或 1 个 credit 包购买。Continue only after at least 21 days and 200 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 400 合格访问 0 API 意向,停止 API 维护。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.

trust-account-delete-brief · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、LinkedIn、小红书与独立开发者社群Planned channels: Google, Bing, LinkedIn, Xiaohongshu.

内容:Content: 围绕 trust-account-delete-brief 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/ASO: create separate exact-intent pages for Apple in-app account deletion, Google Play deletion web resource and multi-system deletion test maps, each linked to current official sources and the canonical tool. OMA/VSO: one weekly English X/GitHub/indie-app scenario card and evidence-led Chinese card only when a traffic slot is granted; never promise approval. GEO: publish a sourced Apple-versus-Google requirement matrix. CRO: the complete free map precedes the $12 pack. Keep X-Paid at zero until organic completions or qualified paid intent appears.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 120 个合格访问30 days · at least 120 qualified visits

继续:Continue: 至少运行 30 天并获得 120 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。Only after at least 30 days and 120 qualified visits, continue if at least 12% complete a deletion map and there is one real purchase or qualified paid intent of at least 3%; at least five completed maps must name two or more real system dependencies and human reviewers must confirm the artifact changed or documented a release decision.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 240 qualified visits after the minimum-age gate, pause active distribution if completion is below 5% with no purchase or qualified paid intent, or if users only need the official platform checklists and do not value a reusable dependency, failure-test and sign-off record. Zero traffic, validation activity and a young launch are not PMF-failure evidence.

trust-form-field-risk · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、LinkedIn、小红书与独立开发者社群Planned channels: Google, Bing, LinkedIn, Xiaohongshu.

内容:Content: 围绕 trust-form-field-risk 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/GEO: publish an exact-intent source page for “data minimization form checklist,” “do I need this form field,” and “privacy review web form,” using sourced one-field examples and clear no-legal-verdict boundaries. OMA/VSO: only after receiving one of three traffic slots, post at most one weekly field-before/after card—English for X/LinkedIn and independently rewritten Chinese for Xiaohongshu—without naming or shaming sites. CRO: full free field review → copied decision brief or second scenario → $7 version-diff register; measure qualified visit → multi-field completion → copy/checkout intent → paid. Partner route: privacy consultants, UX agencies and secure form implementers after five real-form completions. ASO/LBS not applicable. X-Paid stays off until two organic paid intents or one purchase; then exact-intent only at an approved $5/day cap.

频次:Cadence: 每周 1 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits, continue active distribution only if at least 12% complete a multi-field review, at least five users review a real public form rather than a demo, and either qualified paid intent reaches 3% or at least one genuine $7 purchase occurs. Below either time or visit threshold, do not declare failure.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At or after 300 qualified visits, pause active distribution if completion is below 5% with no genuine purchase or qualified paid intent, or users consistently prefer built-in form-builder controls and do not value a separate pre-collection decision record. Close immediately if the tool cannot preserve unknown states, prevent submission data entry or keep legal decisions human-owned.

trust-pixel-purpose-receipt · 等待主动流量席位Waiting for an active traffic slot

Google/Bing 搜索、LinkedIn、小红书与独立开发者社群Planned channels: Google, Bing, LinkedIn, Xiaohongshu.

内容:Content: 围绕 trust-pixel-purpose-receipt 的目标任务,每条内容只展示一个问题、一个决定和可核对证据。OSO/SEO: target the exact-intent terms “pixel purpose owner”, “tag purpose register” and “tracking pixel accountability” with the sourced example already on the page. GEO: expose concise definitions, source links and the distinction between detection evidence and a human purpose decision. OMA/VSO: at most weekly, publish one sanitized orphaned-pixel before/after card to X, LinkedIn and small-agency communities; do not consume an active traffic slot yet. CRO: deliver the free gap receipt first, then offer the $9 editable register and record which scanner/export the visitor already uses. ASO/LBS is not applicable. X-Paid stays off until organic completion or qualified paid intent appears; only then consider a tightly capped exact-intent test with free-scanner and CMP negative terms.

频次:Cadence: 每周 2 个场景Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 150 个合格访问30 days · at least 150 qualified visits

继续:Continue: 至少运行 30 天并获得 150 个合格访问后,只有使用、意向或真实付款达到合同门槛才继续。After at least 30 days and 150 qualified visits from people who already hold a tag-manager or scanner inventory, continue only if at least 10% complete a purpose-owner receipt and there is one real purchase or qualified paid intent of at least 3%; separately record the existing scanner/export and whether a repeat review or change diff occurs.

停止:Stop: 样本成熟后若完成、意向和付款仍低于预设阈值,则暂停渠道或产品;零流量和刚上线不能判失败。At 300 qualified visits, stop if fewer than 5% complete the receipt with no purchase or qualified paid intent, or if interviews and usage show buyers only want automatic scanning, consent enforcement, enterprise governance or legal certification already supplied by stronger products.

vendor-exit · 先验关闭Closed by prior

LinkedIn、采购社群、AI 创业者、小红书Planned channels: LinkedIn, Xiaohongshu.

内容:Content: 用续约前的导出/替代/回归演练讲退出成本。For vendor-exit, each asset shows one target problem, one decision and checkable evidence.

频次:Cadence: 每周 1 个退出清单案例Run on an app-specific cadence; publishing caps are ceilings, not targets. · 30 天 · 至少 100 个合格访问30 days · at least 100 qualified visits

继续:Continue: 每 100 访问 ≥1 合格迁移线索且伙伴接受。Continue only after at least 30 days and 100 qualified visits when use, intent or settled payment reaches the contract.

停止:Stop: 200 访问 0 线索或无伙伴,停止。After the sample matures, pause the channel or product if completion, intent and payment remain below the preset floor; zero traffic or freshness is not failure.