[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AVNH9CCB47FM5Y5NFYCC66":3,"skill-same:01M7AVNH9CCB47FM5Y5NFYCC66":122},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":32,"updatedAt":33,"install":34,"resources":47},"01M7AVNH9CCB47FM5Y5NFYCC66","product-metrics",null,"设计North Star、输入、产品健康、AI质量与商业指标，明确公式、分母、时间窗、数据源、Owner、阈值、护栏和复盘节奏。用于指标体系、Dashboard、实验或发布Gate设计；不在数据不可得时制造虚假精确目标。","bonc-ai\u002Fcogseed","bonc-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F313335383?v=4","01M7AV826PEZKTFGW9MCPBERA7","东方国信 BONC","01M7AV827DEZPMD30Y94T1ZFKQ",19,"MIT","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: product-metrics\ndescription: \"设计North Star、输入、产品健康、AI质量与商业指标，明确公式、分母、时间窗、数据源、Owner、阈值、护栏和复盘节奏。用于指标体系、Dashboard、实验或发布Gate设计；不在数据不可得时制造虚假精确目标。\"\n---\n\n# 产品指标设计\n\n## 目标\n\n把产品价值主张转成可计算、可解释、可行动并能防止局部优化的指标体系。\n\n## 开始前\n\n至少确认：\n\n- 用户价值与业务模式\n- 核心用户旅程和产品阶段\n- 现有数据源、埋点、目标和约束\n\n缺少会改变结论的关键输入时，提出最少必要问题；若用户要求继续，使用 `未知\u002FTBD`，不要补造。\n\n## 工作流\n\n1. 明确指标服务的决策、对象、产品阶段与评审节奏；先检查数据可得性。\n2. 提出单一North Star候选并用客户价值、可理解、可行动、可持续、愿景匹配、可量化和领先性检验。\n3. 构建四层：North Star；3–5个输入指标；质量\u002F留存\u002F可靠性\u002F信任安全健康指标；收入、成本或续费商业指标。\n4. 为每个指标写指标卡：名称、问题、公式、分子、分母、时间窗、去重、分群、数据源、Owner、刷新频率和已知偏差。\n5. 定义基线、目标、预警阈值、护栏、反作弊和分群切片；说明阈值证据或标TBD。\n6. 将指标连接到实验、版本和发布Gate，建立异常→诊断→行动→复盘闭环。\n\n## 专业判断规则\n\n- North Star是客户获得价值的单一指标，不等于收入、OKR或指标列表。\n- 活动量不自动等于价值；必须说明价值代理关系。\n- 平均值必须配合关键分群，防止掩盖弱势用户。\n- AI产品同时跟踪任务成功、采纳\u002F纠正、引用\u002F安全、延迟和成本。\n\n## 输出契约\n\n必须返回：\n\n- 指标树\u002F指标分层\n- 完整指标字典\n- 数据可得性与埋点缺口\n- 目标\u002F阈值\u002F护栏\n- 实验与发布Gate映射\n\n所有关键结论都标为 `事实 \u002F 推断 \u002F 假设 \u002F 建议 \u002F 待决策` 之一，并保留证据ID或缺口。只返回正式资产候选，不直接覆盖PRD、路线图或长期认知资产。\n\n## 失败与降级\n\n- 来源不可访问：列出已完成部分、缺失来源和对结论的影响。\n- 权限或敏感度不满足：停止对应读取或外部动作，返回所需授权。\n- 数据互相冲突：并列冲突及适用条件，不自行裁决为单一事实。\n- 预算耗尽：返回当前证据、未完成步骤和可安全续跑点。\n\n\u003C!-- SKILL-GATE:BEGIN -->\n## Skill Gate 契约\n\n- `use_when`：需要“把产品价值主张转成可计算、可解释、可行动并能防止局部优化的指标体系。”，并已提供或授权“用户价值与业务模式”与“核心用户旅程和产品阶段”等最小业务输入。\n- `do_not_use_when`：缺少或未授权“用户价值与业务模式”；“核心用户旅程和产品阶段”的对象、范围或版本无法确认；任务不属于“product-metrics”职责；或请求违反专属判断规则“North Star是客户获得价值的单一指标，不等于收入、OKR或指标列表。”。通用安全红线仍适用：不得越权、伪造证据或直接覆盖正式资产。\n- `positive_examples`：`请基于已授权材料执行product-metrics，输出指标树\u002F指标分层并保留证据定位。`\n- `negative_examples`：`没有用户价值与业务模式，仍请直接完成product-metrics。`；`核心用户旅程和产品阶段尚未确认，但请直接定稿指标树\u002F指标分层。`\n\n本 Skill 是 `EndUseSkill · L5 · Full · sub_skill · interpreted` 的共享候选能力。自动化晋升天花板为 `staged`，`production_release_allowed: false`。它只产生候选交付物，不执行生产发布。\n\n执行时按需读取以下一层引用：\n\n- 输入\u002F输出和运行边界：[schemas.json](schemas.json)、[references\u002Finput-contract.md](references\u002Finput-contract.md)、[references\u002Foutput-contract.md](references\u002Foutput-contract.md)\n- 本体、验证和失败归因：[references\u002Fontology-mapping.md](references\u002Fontology-mapping.md)、[references\u002Fvalidation-contract.md](references\u002Fvalidation-contract.md)、[references\u002Ffailure-modes.md](references\u002Ffailure-modes.md)\n- 评测、演进和治理：[evals\u002Fevals.json](evals\u002Fevals.json)、[references\u002Fkstar-evolution.md](references\u002Fkstar-evolution.md)、[references\u002Fgovernance-boundaries.md](references\u002Fgovernance-boundaries.md)\n\u003C!-- SKILL-GATE:END -->\n\n## 详细方法\n\n执行前读取 [references\u002Fmethod.md](references\u002Fmethod.md)。需要输出模板、质量Gate和示例时，以该文件为准。\n\n## 来源与适配\n\n主体适配自 phuryn Product Metrics Dashboard 与 North Star Metric；增加B端\u002FAI质量、口径治理、反作弊和Gate映射。 文件级来源、许可证、SHA和修改记录见包根目录 `provenance\u002Fcomponents.yaml` 与 `THIRD_PARTY_NOTICES.md`。\n","# 产品指标设计\n\n## 目标\n\n把产品价值主张转成可计算、可解释、可行动并能防止局部优化的指标体系。\n\n## 开始前\n\n至少确认：\n\n- 用户价值与业务模式\n- 核心用户旅程和产品阶段\n- 现有数据源、埋点、目标和约束\n\n缺少会改变结论的关键输入时，提出最少必要问题；若用户要求继续，使用 `未知\u002FTBD`，不要补造。\n\n## 工作流\n\n1. 明确指标服务的决策、对象、产品阶段与评审节奏；先检查数据可得性。\n2. 提出单一North Star候选并用客户价值、可理解、可行动、可持续、愿景匹配、可量化和领先性检验。\n3. 构建四层：North Star；3–5个输入指标；质量\u002F留存\u002F可靠性\u002F信任安全健康指标；收入、成本或续费商业指标。\n4. 为每个指标写指标卡：名称、问题、公式、分子、分母、时间窗、去重、分群、数据源、Owner、刷新频率和已知偏差。\n5. 定义基线、目标、预警阈值、护栏、反作弊和分群切片；说明阈值证据或标TBD。\n6. 将指标连接到实验、版本和发布Gate，建立异常→诊断→行动→复盘闭环。\n\n## 专业判断规则\n\n- North Star是客户获得价值的单一指标，不等于收入、OKR或指标列表。\n- 活动量不自动等于价值；必须说明价值代理关系。\n- 平均值必须配合关键分群，防止掩盖弱势用户。\n- AI产品同时跟踪任务成功、采纳\u002F纠正、引用\u002F安全、延迟和成本。\n\n## 输出契约\n\n必须返回：\n\n- 指标树\u002F指标分层\n- 完整指标字典\n- 数据可得性与埋点缺口\n- 目标\u002F阈值\u002F护栏\n- 实验与发布Gate映射\n\n所有关键结论都标为 `事实 \u002F 推断 \u002F 假设 \u002F 建议 \u002F 待决策` 之一，并保留证据ID或缺口。只返回正式资产候选，不直接覆盖PRD、路线图或长期认知资产。\n\n## 失败与降级\n\n- 来源不可访问：列出已完成部分、缺失来源和对结论的影响。\n- 权限或敏感度不满足：停止对应读取或外部动作，返回所需授权。\n- 数据互相冲突：并列冲突及适用条件，不自行裁决为单一事实。\n- 预算耗尽：返回当前证据、未完成步骤和可安全续跑点。\n\n\u003C!-- SKILL-GATE:BEGIN -->\n## Skill Gate 契约\n\n- `use_when`：需要“把产品价值主张转成可计算、可解释、可行动并能防止局部优化的指标体系。”，并已提供或授权“用户价值与业务模式”与“核心用户旅程和产品阶段”等最小业务输入。\n- `do_not_use_when`：缺少或未授权“用户价值与业务模式”；“核心用户旅程和产品阶段”的对象、范围或版本无法确认；任务不属于“product-metrics”职责；或请求违反专属判断规则“North Star是客户获得价值的单一指标，不等于收入、OKR或指标列表。”。通用安全红线仍适用：不得越权、伪造证据或直接覆盖正式资产。\n- `positive_examples`：`请基于已授权材料执行product-metrics，输出指标树\u002F指标分层并保留证据定位。`\n- `negative_examples`：`没有用户价值与业务模式，仍请直接完成product-metrics。`；`核心用户旅程和产品阶段尚未确认，但请直接定稿指标树\u002F指标分层。`\n\n本 Skill 是 `EndUseSkill · L5 · Full · sub_skill · interpreted` 的共享候选能力。自动化晋升天花板为 `staged`，`production_release_allowed: false`。它只产生候选交付物，不执行生产发布。\n\n执行时按需读取以下一层引用：\n\n- 输入\u002F输出和运行边界：[schemas.json](schemas.json)、[references\u002Finput-contract.md](references\u002Finput-contract.md)、[references\u002Foutput-contract.md](references\u002Foutput-contract.md)\n- 本体、验证和失败归因：[references\u002Fontology-mapping.md](references\u002Fontology-mapping.md)、[references\u002Fvalidation-contract.md](references\u002Fvalidation-contract.md)、[references\u002Ffailure-modes.md](references\u002Ffailure-modes.md)\n- 评测、演进和治理：[evals\u002Fevals.json](evals\u002Fevals.json)、[references\u002Fkstar-evolution.md](references\u002Fkstar-evolution.md)、[references\u002Fgovernance-boundaries.md](references\u002Fgovernance-boundaries.md)\n\u003C!-- SKILL-GATE:END -->\n\n## 详细方法\n\n执行前读取 [references\u002Fmethod.md](references\u002Fmethod.md)。需要输出模板、质量Gate和示例时，以该文件为准。\n\n## 来源与适配\n\n主体适配自 phuryn Product Metrics Dashboard 与 North Star Metric；增加B端\u002FAI质量、口径治理、反作弊和Gate映射。 文件级来源、许可证、SHA和修改记录见包根目录 `provenance\u002Fcomponents.yaml` 与 `THIRD_PARTY_NOTICES.md`。\n",{"name":5,"description":7},"main","6afb98b7a438d4b993ba23f3af1494c862d80378","resources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6",10,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed","https:\u002F\u002Fcogseed.ai","https:\u002F\u002Fgithub.com\u002Fbonc-ai","https:\u002F\u002Fwww.bonc.com.cn\u002F","北京东方国信科技股份有限公司开源组织 · Enterprise open source by BONC","2026-11-13 21:46:31",{"source":8,"skillName":5,"commands":35,"prompt":45,"zipAvailable":46},[36,39,42],{"tool":37,"command":38},"npx","npx skills add bonc-ai\u002Fcogseed --skill product-metrics",{"tool":40,"command":41},"bunx","bunx skills add bonc-ai\u002Fcogseed --skill product-metrics",{"tool":43,"command":44},"pnpm","pnpm dlx skills add bonc-ai\u002Fcogseed --skill product-metrics","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002FSKILL.md",true,[48,54,58,62,66,70,74,78,82,86,90,94,98,102,106,110,114,118],{"path":49,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":52,"sourceHtmlUrl":53},"_meta.json","REFERENCE","参考文档",73,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002F_meta.json",{"path":55,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":56,"sourceHtmlUrl":57},"agents\u002Fopenai.yaml",810,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fagents\u002Fopenai.yaml",{"path":59,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":60,"sourceHtmlUrl":61},"evals\u002Fevals.json",13878,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fevals\u002Fevals.json",{"path":63,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":64,"sourceHtmlUrl":65},"evals\u002Fforecast_model.md",251,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fevals\u002Fforecast_model.md",{"path":67,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":68,"sourceHtmlUrl":69},"evals\u002Foutcome_evaluation.md",259,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fevals\u002Foutcome_evaluation.md",{"path":71,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":72,"sourceHtmlUrl":73},"evals\u002Fregression_tests.md",444,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fevals\u002Fregression_tests.md",{"path":75,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":76,"sourceHtmlUrl":77},"evals\u002Freplay_dataset.md",254,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fevals\u002Freplay_dataset.md",{"path":79,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Feval-cases.yaml",13570,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Feval-cases.yaml",{"path":83,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Ffailure-modes.md",1314,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Ffailure-modes.md",{"path":87,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":88,"sourceHtmlUrl":89},"references\u002Fgovernance-boundaries.md",786,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fgovernance-boundaries.md",{"path":91,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":92,"sourceHtmlUrl":93},"references\u002Finput-contract.md",1019,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Finput-contract.md",{"path":95,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":96,"sourceHtmlUrl":97},"references\u002Fkstar-evolution.md",754,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fkstar-evolution.md",{"path":99,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":100,"sourceHtmlUrl":101},"references\u002Fmethod.md",2213,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fmethod.md",{"path":103,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":104,"sourceHtmlUrl":105},"references\u002Fontology-mapping.md",1227,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fontology-mapping.md",{"path":107,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":108,"sourceHtmlUrl":109},"references\u002Foutput-contract.md",1056,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Foutput-contract.md",{"path":111,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":112,"sourceHtmlUrl":113},"references\u002Fskill-spec.yaml",1807,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fskill-spec.yaml",{"path":115,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":116,"sourceHtmlUrl":117},"references\u002Fvalidation-contract.md",1087,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Freferences\u002Fvalidation-contract.md",{"path":119,"resourceType":50,"resourceTypeDesc":51,"sizeBytes":120,"sourceHtmlUrl":121},"schemas.json",3908,"https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fbuiltin\u002Fmarketplace\u002Fskills\u002Fa827416958f6\u002Fschemas.json",[123,130,137,144,151,158],{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":129,"publishedAt":19},"01M7AVTAEJDDZ3YFV85AK28FWN","sample-supply-chain","供应链样本：安装钩子 + 拼写伪装 + 源覆盖。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-supply-chain-skill\u002FSKILL.md",[],[],{"skillNo":131,"name":132,"nameZh":6,"description":133,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":134,"categories":135,"keywords":136,"publishedAt":19},"01M7AVT84YQ5GZDMB9PRTEBV8V","sample-prose-only","纯散文安全守则样本：禁止某行为的说明不等于该行为。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-prose-only-skill\u002FSKILL.md",[],[],{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":141,"categories":142,"keywords":143,"publishedAt":19},"01M7AVT59NWX4M2EZJKRXMEPGB","sample-obfuscated","混淆载荷样本：解码与执行共现。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-obfuscated-skill\u002FSKILL.md",[],[],{"skillNo":145,"name":146,"nameZh":6,"description":147,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":148,"categories":149,"keywords":150,"publishedAt":19},"01M7AVT246DRNSJ5D9JQRK4QKP","sample-hidden-content","隐藏内容载荷样本：用于回归「不可见字符不做上下文降权」这一不变量。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-hidden-content-skill\u002FSKILL.md",[],[],{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":155,"categories":156,"keywords":157,"publishedAt":19},"01M7AVSYR91DFC262VQ23BAZA1","sample-evasion","反规避样本：危险内容只以「不可见字符切断」与「编码后」两种形式出现。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-evasion-skill\u002FSKILL.md",[],[],{"skillNo":159,"name":160,"nameZh":6,"description":161,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":162,"categories":163,"keywords":164,"publishedAt":19},"01M7AVSVMGH458JZC3Q454QWTK","sample-agentic","代理运行时风险样本：配置篡改 + 隐蔽指令。","https:\u002F\u002Fgithub.com\u002Fbonc-ai\u002Fcogseed\u002Fblob\u002Fmain\u002Fresources\u002Fguardrail\u002Fskill-sentry\u002Ftests\u002Ffixtures\u002Fv2-samples\u002Fsample-agentic-skill\u002FSKILL.md",[],[]]