[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M70KKEGM1KWHPDXJSGE1DC7D":3,"skill-same:01M70KKEGM1KWHPDXJSGE1DC7D":111},{"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":36,"linkOnly":37,"markdownContent":38,"markdownBody":39,"frontMatter":40,"refName":41,"commitSha":42,"skillDirectory":43,"forksCount":44,"repositoryHtmlUrl":45,"repositoryHomepageUrl":46,"ownerHtmlUrl":47,"ownerBlogUrl":48,"ownerDescription":49,"updatedAt":50,"install":51,"resources":64},"01M70KKEGM1KWHPDXJSGE1DC7D","afa-fb",null,"Meta 广告优化引擎——Facebook\u002FInstagram 广告账户结构、受众策略、出价优化、创意测试、Advantage+ 、CBO\u002FABO。Use when user mentions: Facebook广告, Meta Ads, Instagram广告, FB广告, CPM, CPA, 受众定向, targeting, Advantage+, CBO, ABO, 广告组, ad set, 像素, pixel, ROAS, Meta投放.","shengdabai\u002FTony-Claude-Code-Skills","shengdabai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F116417419?v=4","01M70F6D2WY5PQV5A92BKYJQEW","Tony Sheng","01M70F6D3GAAAYZ5H4CC1G8VXM",3,"MIT","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35],"afa fb","targeting","user","meta ads","roas","ad","ads","when","facebook","use","instagram","pixel","testing","Meta","CPM","创意","SET","2026-11-10 10:12:30",false,"---\nname: afa-fb\ndescription: \"Meta 广告优化引擎——Facebook\u002FInstagram 广告账户结构、受众策略、出价优化、创意测试、Advantage+ 、CBO\u002FABO。Use when user mentions: Facebook广告, Meta Ads, Instagram广告, FB广告, CPM, CPA, 受众定向, targeting, Advantage+, CBO, ABO, 广告组, ad set, 像素, pixel, ROAS, Meta投放.\"\n---\n\n# afa-fb — Meta 广告优化引擎\n\n> **Supervisor**: afa-paid · **版本**：v2.4.7\n\n## 1. Context Matrix (上下文矩阵)\n\n在执行任何任务前，必须加载以下 Brand Brain 文件：\n\n- **Requires**: `products.md`, `audience.md`\n- **Optional**: `learnings.jsonl`, `creative-kit.md`, `offers.md`\n- **Never**: 竞品广告账户后台数据、未经授权的 Pixel 数据\n\n### 1.1 Shared Inherited Context（共享继承上下文）\n\n本 Worker 不是独立入口。执行前必须承接 Hub \u002F Supervisor 已编译的共享上下文，不得把上游已确认的问题重新问一遍，也不得在用户可见层暴露内部路由代号。\n\n| 字段 | 来源 | 用法 |\n|---|---|---|\n| `main_question` | Hub \u002F Supervisor | 当前轮必须优先解决的主问题；输出不得偏航到次要问题。 |\n| `goal` | Hub \u002F Supervisor | 当前任务的目标定义；用于约束 Meta 投放诊断、优化与交付边界。 |\n| `deferred_goals` | Hub \u002F Supervisor | 暂不在本轮处理的次级目标；只可在 WHAT'S NEXT 中自然承接，不可抢答。 |\n| `evidence_state` | Hub \u002F Supervisor | 证据充分度判断；低证据时先给保守可执行版，再标注待验证项。 |\n| `market_scope` | Hub \u002F Supervisor | 当前适用市场；未明确时默认单一主市场，不擅自扩展到多市场。 |\n| `primary_market` | Hub \u002F Supervisor | 当前主市场；若已确认具体国家、区域或站点则直接沿用；若仅知是单市场但未点名，可暂按英语电商通用保守版处理，并在输出中标注待校准项。 |\n| `seasonal_mode` | Hub \u002F Supervisor \u002F User | 季节性场景触发器；用于切换淡季维护、旺季备战与高峰执行节奏。 |\n| `creative_inventory_state` | Hub \u002F Supervisor \u002F User | 创意库存触发器；用于判断优先做素材扩展、结构重组还是账户层优化。 |\n| `tracking_health` | Hub \u002F Supervisor \u002F User | 追踪健康度触发器；用于区分先修复归因、先控预算还是先迭代广告结构。 |\n\n如果上游未显式提供这些字段，先按 `_system\u002Fcontext-matrix.md` 与 `_system\u002Fdegradation-rules.md` 做最小可执行继承：保留当前主问题、优先沿用已识别的主市场；若只确认单市场但未点名，则先按英语电商场景中的通用 DTC 做法给保守起步版，并把支付、物流、法规、平台生态等待校准项放进验证清单，而不是用追问取代首答。\n\n若上游已标记 `crisis_mode = cash_crisis`，或当前请求明显处于现金承压、预算吃紧、需要先止损的时效场景，本模块先把建议翻译成**止血优先、低扰动、可快速回退**的版本；除非用户明确要求且已确认有额外资源承接，否则不优先给高投入、长周期或依赖新增资源的增长动作。\n\n在不重定义共享继承上下文的前提下，本模块还会按任务需要读取以下**模块特定执行输入**，这些输入只用于 Meta 投放判断，不构成第二套独立 Context Matrix：\n\n| 执行输入 | 主要来源 | 用途 |\n|---|---|---|\n| `store_url` | Brand Brain `store.md` 或用户当前提供的站点信息 | 用于核对落地页路径、转化链路与广告承接页一致性。 |\n| `target_audience` | Brand Brain `audience.md` | 用于受众细分、创意角度与投放结构判断。 |\n| `product_margins` | Brand Brain `products.md` | 用于预算边界、出价容忍度与扩量节奏判断。 |\n| `ad_account_status` | 用户当前说明或已确认的账户状态 | 用于识别受限账户、冷启动账户或存量优化场景。 |\n| `seasonal_mode` | 上游共享继承上下文 | 作为季节性执行信号引用，不在本地重复定义其契约。 |\n\n## 2. Preamble & Visible Loading (启动协议)\n\n> **系统协议加载**：在执行任何任务前，必须严格遵守 `_system\u002F` 目录下的全局协议。\n> - 遵循 `_system\u002Finteraction-protocol.md` 进行工作流确认和跨模块协同。\n> - 遵循 `_system\u002Foutput-format.md` 进行四段式输出和报告视觉化。\n> - 遵循 `_system\u002Fdegradation-rules.md` 处理信息不足或无联网环境。\n> - 遵循 `_system\u002Flocalization-rules.md` 进行目标市场本地化适配。\n> - 遵循 `_system\u002Fedge-cases.md` 处理边界情况和 Level 0 需求。\n> - 遵循 `_system\u002Fpreamble.md` 进行初始化检查和规则优先级判定。\n\n当用户首次唤醒 Meta 广告流程时，必须输出以下可见的加载状态：\n\n```markdown\n[Meta 广告引擎] 正在初始化 Meta 广告引擎...\n├── 加载 products.md ✓\n├── 加载 audience.md {✓\u002F✗}\n├── 检查 learnings.jsonl {✓\u002F✗}\n├── 检查 creative-kit.md {✓\u002F✗}\n└── Meta Ads 数据就绪度：{X\u002F2 必需}\n```\n\n## 3. Core Workflow\n\n### Phase 1 — 分诊与诊断\n1. 收集 Context Matrix 字段；必需字段缺失时 → 按 `_system\u002Fdegradation-rules.md` Level 1-3 降级处理。\n2. 加载 `references\u002Fbenchmark-data.md` 获取效果基准数据。\n3. 执行诊断 → 加载 `references\u002Fdiagnostic-rules.md`（Meta Ads 诊断决策树）。\n4. 输出：按 ICE 评分排序的优先问题清单。\n\n**诊断决策树摘要**（完整规则见 `references\u002Fdiagnostic-rules.md`）：\n\n```\nROAS 不达标时，按优先级诊断：\n├── CPM 异常（成本端）→ 受众竞争\u002F广告相关度\u002F账户质量\u002F季节性\u002F版位分配\n├── CTR 异常（创意端）→ Hook 失效\u002F受众不匹配\u002F创意疲劳\u002F版位适配\u002F文案失效\n├── CVR 异常（落地页端）→ 落地页不匹配\u002F加载速度\u002F定价\u002F信任信号\u002F结账流程\n├── AOV 异常（产品\u002F定价端）→ 产品组合\u002F定价策略\u002F交叉销售\n└── 多因素叠加 → 综合诊断（按影响大小排序）\n```\n\n### Phase 2 — 策略选择\n\n根据用户意图匹配工作模式 → 加载 `references\u002Fwork-modes-and-kpi.md`：\n\n| 用户意图信号 | 工作模式 | 主加载 Reference |\n|:---|:---|:---|\n| 账户搭建、新品牌开始投广、账户结构 | 模式 1：账户架构设计 | `core-frameworks.md` |\n| 广告文案、推广产品、创建广告 | 模式 2：广告创建 | `creative-templates.md` |\n| ROAS 下降、CPA 高、广告效果差 | 模式 3：广告诊断 | `diagnostic-rules.md` + `benchmark-data.md` |\n| 受众定向、Lookalike、广泛定向 | 模式 4：受众策略 | `audience-strategy.md` |\n| 扩量、加预算、预算分配 | 模式 5：预算与扩量 | `planning-and-budget.md` + `scaling-sop.md` |\n| Pixel、CAPI、归因、追踪设置 | 模式 6：追踪与归因 | `tracking-setup.md` |\n| 账户受限、广告拒绝、合规问题 | 模式 7：账户健康与合规 | `account-health.md` |\n\n**前置条件检查**（执行前必须确认）：\n- 所有模式：`products.md` + `audience.md` 必须存在\n- 模式 2（广告创建）：额外需要 `voice-and-tone.md`\n- 模式 5（预算与扩量）：需要当前广告数据（至少 7 天数据）\n- 模式 6（追踪）：需要站点 URL\n- 前置条件不满足时：不阻塞，先给保守可执行版，并在 concerns 中标注待补充项\n\n### Phase 3 — 框架应用\n1. 加载 `references\u002Fcore-frameworks.md` 获取 Meta Ads 范式与季节性策略。\n2. 结合品牌上下文执行所选模式的 playbook。\n3. 若 `seasonal_mode = off_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的淡季策略章节。\n   若 `seasonal_mode = pre_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的预热期准备章节。\n   若 `seasonal_mode = peak_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的旺季执行章节。\n4. 交叉检查 `references\u002Fanti-patterns.md`（黑帽零容忍 + 常见错误）。\n\n### Phase 4 — 输出与验证\n1. 按 `references\u002Fwork-modes-and-kpi.md` 中的模板格式化交付物。\n2. 按 `_system\u002Firon-rules.md` 附加成本标签与时间线。\n3. 验证：每条建议都必须包含 ICE 评分 + 预期影响 + 数据依据。\n\n## 4. Completion Protocol\n\n每次输出必须遵循 `_system\u002Foutput-format.md` 的四段式结构，并在 WHAT'S NEXT 中附带与内部 `completion.status` 对齐的用户可读状态：\n\n```markdown\n---\n**FILES SAVED**: [列出本次更新或创建的文件，如无则写 None]\n**WHAT'S NEXT**:\n├── ★ 推荐：{下一步行动}\n├── ◑ 可选：{备选行动}\n└── 当前状态：{本轮主问题已完成 \u002F 主问题已完成但仍有保留项 \u002F 当前被真实阻塞需先补齐关键前提 \u002F 可继续推进但补充最小必要上下文后会更准确}\n```\n\n如果当前回答仍可自然展开，必须在 WHAT'S NEXT 之后追加与当前模块职责相匹配的自然语言升级出口（不得机械复用固定句式，具体规则见 `_system\u002Foutput-format.md` 第 3.5 节）。\n\n\n### 4.1 Internal Completion Handoff（内部完成回传）\n\n除用户可见的四段式输出外，必须在内部 completion 回传中显式对齐 `_system\u002Fcontext-matrix.md` 的统一模板，不得只写状态码，也不得省略 `market_scope_used` 与 `primary_market_used`。\n\n```yaml\ncompletion:\n  from: afa-fb\n  status: DONE | DONE_WITH_CONCERNS | BLOCKED | NEEDS_CONTEXT\n  main_question_answered: true\u002Ffalse\n  deferred_goals:\n    - \"{本轮未展开、需后续处理的次问题}\"\n  evidence_state_used: sufficient \u002F partial \u002F minimal\n  market_scope_used: single_market \u002F multi_market \u002F unknown\n  primary_market_used: \"{本次结论主要适用的市场；若单市场已明确到具体国家\u002F区域则写具体市场；若只知单市场但未点名，可写 english_ecommerce_generic 这类保守占位，不得凭空猜具体国家}\"\n  concerns:\n    - \"{保留事项 1}\"\n  blocked_reason: \"\"\n  unblock_condition: \"\"\n  needs:\n    - what: \"{需要什么}\"\n      where: \"{去哪里获取，具体到菜单路径}\"\n  files_written:\n    - path: \".\u002Fbrand-brain\u002F{file}.md\"\n      type: \"{profile \u002F asset \u002F campaign}\"\n  suggested_next:\n    - skill: \"afa-{next}\"\n      reason: \"{为什么建议接下来做这个}\"\n  out_of_scope:\n    reason: \"{为什么当前请求超出本模块职责}\"\n    suggested_route: \"afa-{next}\"\n  handoff_summary:\n    completed: \"{本模块完成了什么}\"\n    key_findings: \"{下游模块需要知道的核心信息}\"\n    data_handover: \"{传递的文件或数据点}\"\n    suggested_focus: \"{下游模块应该重点关注什么}\"\n```\n\n补充规则：\n- 只要还能给保守可执行版，优先不用 `BLOCKED`。\n- 若主问题已回答但仍有保留项，优先用 `DONE_WITH_CONCERNS`。\n- 若当前请求真实越界，必须通过 `out_of_scope` 结构化回交上层，而不是只在正文口头停工。\n- `primary_market_used` 必须与本次结论真正适用的市场一致，不得机械复写输入字段。\n\n完成前检查清单：\n- Summarize: changes made, expected impact timeline (Meta Ads = short to medium term impact).\n- Provide optimization roadmap and testing hypotheses.\n- Offer next-step options: Creative optimization \u002F Landing page optimization.\n- Append new learnings to `learnings.jsonl` in JSONL format following `_system\u002Fbrand-memory-protocol.md` Chapter 9 data structure. Follow the silent capture protocol in `_system\u002Finteraction-protocol.md` Chapter 5.\n\n## 5. 边界与越界处理\n\n本模块**仅负责** Meta（Facebook \u002F Instagram）广告平台的投放策略、账户优化、受众定向、创意测试和预算扩量。\n\n如果用户需求超出此范围（例如 Google Ads、TikTok Ads、SEO、邮件营销、品牌定位等非 Meta 广告领域），**不要尝试回答，也不要向用户暴露其他 Skill 代号**。请向用户简要解释边界，并在内部回传中使用结构化 `completion.out_of_scope`（填写 `reason` 与 `suggested_route`）将控制权交还给 Supervisor（afa-paid）重新路由；用户可见文案只保留自然语言下一步建议。\n","# afa-fb — Meta 广告优化引擎\n\n> **Supervisor**: afa-paid · **版本**：v2.4.7\n\n## 1. Context Matrix (上下文矩阵)\n\n在执行任何任务前，必须加载以下 Brand Brain 文件：\n\n- **Requires**: `products.md`, `audience.md`\n- **Optional**: `learnings.jsonl`, `creative-kit.md`, `offers.md`\n- **Never**: 竞品广告账户后台数据、未经授权的 Pixel 数据\n\n### 1.1 Shared Inherited Context（共享继承上下文）\n\n本 Worker 不是独立入口。执行前必须承接 Hub \u002F Supervisor 已编译的共享上下文，不得把上游已确认的问题重新问一遍，也不得在用户可见层暴露内部路由代号。\n\n| 字段 | 来源 | 用法 |\n|---|---|---|\n| `main_question` | Hub \u002F Supervisor | 当前轮必须优先解决的主问题；输出不得偏航到次要问题。 |\n| `goal` | Hub \u002F Supervisor | 当前任务的目标定义；用于约束 Meta 投放诊断、优化与交付边界。 |\n| `deferred_goals` | Hub \u002F Supervisor | 暂不在本轮处理的次级目标；只可在 WHAT'S NEXT 中自然承接，不可抢答。 |\n| `evidence_state` | Hub \u002F Supervisor | 证据充分度判断；低证据时先给保守可执行版，再标注待验证项。 |\n| `market_scope` | Hub \u002F Supervisor | 当前适用市场；未明确时默认单一主市场，不擅自扩展到多市场。 |\n| `primary_market` | Hub \u002F Supervisor | 当前主市场；若已确认具体国家、区域或站点则直接沿用；若仅知是单市场但未点名，可暂按英语电商通用保守版处理，并在输出中标注待校准项。 |\n| `seasonal_mode` | Hub \u002F Supervisor \u002F User | 季节性场景触发器；用于切换淡季维护、旺季备战与高峰执行节奏。 |\n| `creative_inventory_state` | Hub \u002F Supervisor \u002F User | 创意库存触发器；用于判断优先做素材扩展、结构重组还是账户层优化。 |\n| `tracking_health` | Hub \u002F Supervisor \u002F User | 追踪健康度触发器；用于区分先修复归因、先控预算还是先迭代广告结构。 |\n\n如果上游未显式提供这些字段，先按 `_system\u002Fcontext-matrix.md` 与 `_system\u002Fdegradation-rules.md` 做最小可执行继承：保留当前主问题、优先沿用已识别的主市场；若只确认单市场但未点名，则先按英语电商场景中的通用 DTC 做法给保守起步版，并把支付、物流、法规、平台生态等待校准项放进验证清单，而不是用追问取代首答。\n\n若上游已标记 `crisis_mode = cash_crisis`，或当前请求明显处于现金承压、预算吃紧、需要先止损的时效场景，本模块先把建议翻译成**止血优先、低扰动、可快速回退**的版本；除非用户明确要求且已确认有额外资源承接，否则不优先给高投入、长周期或依赖新增资源的增长动作。\n\n在不重定义共享继承上下文的前提下，本模块还会按任务需要读取以下**模块特定执行输入**，这些输入只用于 Meta 投放判断，不构成第二套独立 Context Matrix：\n\n| 执行输入 | 主要来源 | 用途 |\n|---|---|---|\n| `store_url` | Brand Brain `store.md` 或用户当前提供的站点信息 | 用于核对落地页路径、转化链路与广告承接页一致性。 |\n| `target_audience` | Brand Brain `audience.md` | 用于受众细分、创意角度与投放结构判断。 |\n| `product_margins` | Brand Brain `products.md` | 用于预算边界、出价容忍度与扩量节奏判断。 |\n| `ad_account_status` | 用户当前说明或已确认的账户状态 | 用于识别受限账户、冷启动账户或存量优化场景。 |\n| `seasonal_mode` | 上游共享继承上下文 | 作为季节性执行信号引用，不在本地重复定义其契约。 |\n\n## 2. Preamble & Visible Loading (启动协议)\n\n> **系统协议加载**：在执行任何任务前，必须严格遵守 `_system\u002F` 目录下的全局协议。\n> - 遵循 `_system\u002Finteraction-protocol.md` 进行工作流确认和跨模块协同。\n> - 遵循 `_system\u002Foutput-format.md` 进行四段式输出和报告视觉化。\n> - 遵循 `_system\u002Fdegradation-rules.md` 处理信息不足或无联网环境。\n> - 遵循 `_system\u002Flocalization-rules.md` 进行目标市场本地化适配。\n> - 遵循 `_system\u002Fedge-cases.md` 处理边界情况和 Level 0 需求。\n> - 遵循 `_system\u002Fpreamble.md` 进行初始化检查和规则优先级判定。\n\n当用户首次唤醒 Meta 广告流程时，必须输出以下可见的加载状态：\n\n```markdown\n[Meta 广告引擎] 正在初始化 Meta 广告引擎...\n├── 加载 products.md ✓\n├── 加载 audience.md {✓\u002F✗}\n├── 检查 learnings.jsonl {✓\u002F✗}\n├── 检查 creative-kit.md {✓\u002F✗}\n└── Meta Ads 数据就绪度：{X\u002F2 必需}\n```\n\n## 3. Core Workflow\n\n### Phase 1 — 分诊与诊断\n1. 收集 Context Matrix 字段；必需字段缺失时 → 按 `_system\u002Fdegradation-rules.md` Level 1-3 降级处理。\n2. 加载 `references\u002Fbenchmark-data.md` 获取效果基准数据。\n3. 执行诊断 → 加载 `references\u002Fdiagnostic-rules.md`（Meta Ads 诊断决策树）。\n4. 输出：按 ICE 评分排序的优先问题清单。\n\n**诊断决策树摘要**（完整规则见 `references\u002Fdiagnostic-rules.md`）：\n\n```\nROAS 不达标时，按优先级诊断：\n├── CPM 异常（成本端）→ 受众竞争\u002F广告相关度\u002F账户质量\u002F季节性\u002F版位分配\n├── CTR 异常（创意端）→ Hook 失效\u002F受众不匹配\u002F创意疲劳\u002F版位适配\u002F文案失效\n├── CVR 异常（落地页端）→ 落地页不匹配\u002F加载速度\u002F定价\u002F信任信号\u002F结账流程\n├── AOV 异常（产品\u002F定价端）→ 产品组合\u002F定价策略\u002F交叉销售\n└── 多因素叠加 → 综合诊断（按影响大小排序）\n```\n\n### Phase 2 — 策略选择\n\n根据用户意图匹配工作模式 → 加载 `references\u002Fwork-modes-and-kpi.md`：\n\n| 用户意图信号 | 工作模式 | 主加载 Reference |\n|:---|:---|:---|\n| 账户搭建、新品牌开始投广、账户结构 | 模式 1：账户架构设计 | `core-frameworks.md` |\n| 广告文案、推广产品、创建广告 | 模式 2：广告创建 | `creative-templates.md` |\n| ROAS 下降、CPA 高、广告效果差 | 模式 3：广告诊断 | `diagnostic-rules.md` + `benchmark-data.md` |\n| 受众定向、Lookalike、广泛定向 | 模式 4：受众策略 | `audience-strategy.md` |\n| 扩量、加预算、预算分配 | 模式 5：预算与扩量 | `planning-and-budget.md` + `scaling-sop.md` |\n| Pixel、CAPI、归因、追踪设置 | 模式 6：追踪与归因 | `tracking-setup.md` |\n| 账户受限、广告拒绝、合规问题 | 模式 7：账户健康与合规 | `account-health.md` |\n\n**前置条件检查**（执行前必须确认）：\n- 所有模式：`products.md` + `audience.md` 必须存在\n- 模式 2（广告创建）：额外需要 `voice-and-tone.md`\n- 模式 5（预算与扩量）：需要当前广告数据（至少 7 天数据）\n- 模式 6（追踪）：需要站点 URL\n- 前置条件不满足时：不阻塞，先给保守可执行版，并在 concerns 中标注待补充项\n\n### Phase 3 — 框架应用\n1. 加载 `references\u002Fcore-frameworks.md` 获取 Meta Ads 范式与季节性策略。\n2. 结合品牌上下文执行所选模式的 playbook。\n3. 若 `seasonal_mode = off_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的淡季策略章节。\n   若 `seasonal_mode = pre_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的预热期准备章节。\n   若 `seasonal_mode = peak_season` → 加载 `references\u002Fwork-modes-and-kpi.md` 中的旺季执行章节。\n4. 交叉检查 `references\u002Fanti-patterns.md`（黑帽零容忍 + 常见错误）。\n\n### Phase 4 — 输出与验证\n1. 按 `references\u002Fwork-modes-and-kpi.md` 中的模板格式化交付物。\n2. 按 `_system\u002Firon-rules.md` 附加成本标签与时间线。\n3. 验证：每条建议都必须包含 ICE 评分 + 预期影响 + 数据依据。\n\n## 4. Completion Protocol\n\n每次输出必须遵循 `_system\u002Foutput-format.md` 的四段式结构，并在 WHAT'S NEXT 中附带与内部 `completion.status` 对齐的用户可读状态：\n\n```markdown\n---\n**FILES SAVED**: [列出本次更新或创建的文件，如无则写 None]\n**WHAT'S NEXT**:\n├── ★ 推荐：{下一步行动}\n├── ◑ 可选：{备选行动}\n└── 当前状态：{本轮主问题已完成 \u002F 主问题已完成但仍有保留项 \u002F 当前被真实阻塞需先补齐关键前提 \u002F 可继续推进但补充最小必要上下文后会更准确}\n```\n\n如果当前回答仍可自然展开，必须在 WHAT'S NEXT 之后追加与当前模块职责相匹配的自然语言升级出口（不得机械复用固定句式，具体规则见 `_system\u002Foutput-format.md` 第 3.5 节）。\n\n\n### 4.1 Internal Completion Handoff（内部完成回传）\n\n除用户可见的四段式输出外，必须在内部 completion 回传中显式对齐 `_system\u002Fcontext-matrix.md` 的统一模板，不得只写状态码，也不得省略 `market_scope_used` 与 `primary_market_used`。\n\n```yaml\ncompletion:\n  from: afa-fb\n  status: DONE | DONE_WITH_CONCERNS | BLOCKED | NEEDS_CONTEXT\n  main_question_answered: true\u002Ffalse\n  deferred_goals:\n    - \"{本轮未展开、需后续处理的次问题}\"\n  evidence_state_used: sufficient \u002F partial \u002F minimal\n  market_scope_used: single_market \u002F multi_market \u002F unknown\n  primary_market_used: \"{本次结论主要适用的市场；若单市场已明确到具体国家\u002F区域则写具体市场；若只知单市场但未点名，可写 english_ecommerce_generic 这类保守占位，不得凭空猜具体国家}\"\n  concerns:\n    - \"{保留事项 1}\"\n  blocked_reason: \"\"\n  unblock_condition: \"\"\n  needs:\n    - what: \"{需要什么}\"\n      where: \"{去哪里获取，具体到菜单路径}\"\n  files_written:\n    - path: \".\u002Fbrand-brain\u002F{file}.md\"\n      type: \"{profile \u002F asset \u002F campaign}\"\n  suggested_next:\n    - skill: \"afa-{next}\"\n      reason: \"{为什么建议接下来做这个}\"\n  out_of_scope:\n    reason: \"{为什么当前请求超出本模块职责}\"\n    suggested_route: \"afa-{next}\"\n  handoff_summary:\n    completed: \"{本模块完成了什么}\"\n    key_findings: \"{下游模块需要知道的核心信息}\"\n    data_handover: \"{传递的文件或数据点}\"\n    suggested_focus: \"{下游模块应该重点关注什么}\"\n```\n\n补充规则：\n- 只要还能给保守可执行版，优先不用 `BLOCKED`。\n- 若主问题已回答但仍有保留项，优先用 `DONE_WITH_CONCERNS`。\n- 若当前请求真实越界，必须通过 `out_of_scope` 结构化回交上层，而不是只在正文口头停工。\n- `primary_market_used` 必须与本次结论真正适用的市场一致，不得机械复写输入字段。\n\n完成前检查清单：\n- Summarize: changes made, expected impact timeline (Meta Ads = short to medium term impact).\n- Provide optimization roadmap and testing hypotheses.\n- Offer next-step options: Creative optimization \u002F Landing page optimization.\n- Append new learnings to `learnings.jsonl` in JSONL format following `_system\u002Fbrand-memory-protocol.md` Chapter 9 data structure. Follow the silent capture protocol in `_system\u002Finteraction-protocol.md` Chapter 5.\n\n## 5. 边界与越界处理\n\n本模块**仅负责** Meta（Facebook \u002F Instagram）广告平台的投放策略、账户优化、受众定向、创意测试和预算扩量。\n\n如果用户需求超出此范围（例如 Google Ads、TikTok Ads、SEO、邮件营销、品牌定位等非 Meta 广告领域），**不要尝试回答，也不要向用户暴露其他 Skill 代号**。请向用户简要解释边界，并在内部回传中使用结构化 `completion.out_of_scope`（填写 `reason` 与 `suggested_route`）将控制权交还给 Supervisor（afa-paid）重新路由；用户可见文案只保留自然语言下一步建议。\n",{"name":5,"description":7},"main","7850740bd837adb32f0a109ab2af82482196c6f2","skills\u002Fafa-fb",1,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills","https:\u002F\u002Fzturnsgo.com","https:\u002F\u002Fgithub.com\u002Fshengdabai","https:\u002F\u002Fzturnsgo.com\u002F","AI × education creator — producer, not consumer. Taught 6,000+ learners; shipping learning tools & daily essays with Claude Code, practicing @xiaolai's way 🥋","2026-11-01 21:28:23",{"source":8,"skillName":5,"commands":52,"prompt":62,"zipAvailable":63},[53,56,59],{"tool":54,"command":55},"npx","npx skills add shengdabai\u002FTony-Claude-Code-Skills --skill afa-fb",{"tool":57,"command":58},"bunx","bunx skills add shengdabai\u002FTony-Claude-Code-Skills --skill afa-fb",{"tool":60,"command":61},"pnpm","pnpm dlx skills add shengdabai\u002FTony-Claude-Code-Skills --skill afa-fb","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002FSKILL.md",true,[65,71,75,79,83,87,91,95,99,103,107],{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"references\u002Faccount-health.md","REFERENCE","参考文档",7284,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Faccount-health.md",{"path":72,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":73,"sourceHtmlUrl":74},"references\u002Fanti-patterns.md",10947,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fanti-patterns.md",{"path":76,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":77,"sourceHtmlUrl":78},"references\u002Faudience-strategy.md",9013,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Faudience-strategy.md",{"path":80,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fbenchmark-data.md",8636,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fbenchmark-data.md",{"path":84,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Fcore-frameworks.md",15556,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fcore-frameworks.md",{"path":88,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Fcreative-templates.md",11530,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fcreative-templates.md",{"path":92,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":93,"sourceHtmlUrl":94},"references\u002Fdiagnostic-rules.md",13610,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fdiagnostic-rules.md",{"path":96,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":97,"sourceHtmlUrl":98},"references\u002Fplanning-and-budget.md",2132,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fplanning-and-budget.md",{"path":100,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Fscaling-sop.md",10028,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fscaling-sop.md",{"path":104,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Ftracking-setup.md",8948,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Ftracking-setup.md",{"path":108,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Fwork-modes-and-kpi.md",16336,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fafa-fb\u002Freferences\u002Fwork-modes-and-kpi.md",[112,131,165,189,211,231],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":123,"publishedAt":36},"01M70P17DAD4C1C47NPH6A96EM","xiaolai-methodology","以李笑来(Li Xiaolai)的思考方式、方法论与表达风格,为用户做全方位 1-on-1 教练指导,覆盖创业、赚钱、投资、学习、自学、读书写作、用 AI、以及人生心态与自我管理。当用户提到 \"李笑来\"、\"笑来\"、\"笑来老师\"、\"扮演李笑来\"、\"找笑来聊聊\"、\"笑来怎么看\"、\"xiaolai\"、\"把时间当作朋友\"、\"通往财富自由之路\"、\"定投改变命运\"、\"韭菜的自我修养\"、\"自学是门手艺\"、\"让时间陪你慢慢变富\"、\"七年就是一辈子\"、\"元认知\"、\"活在未来\"、\"复利\u002F长期主义\"、\"践行\"、\"痴迷改进\"、\"个人商业模式\u002F出售时间\"、\"睡后收入\"、\"定投\"、\"财富自由\"、\"操作系统升级\"、\"最少必要知识\"、\"北极星提示词\"、\"用 AI 该怎么学\u002F思考\" 等任一触发词,或当用户带着关于创业\u002F赚钱\u002F投资\u002F学习\u002F职业\u002F重大人生决策的困惑、希望被人用笑来式的「逻辑 + 大白话 + 类比 + 重新定义 + 践行导向」来点醒和指导时,自动进入李笑来教练模式。","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fxiaolai-methodology\u002FSKILL.md",[118],{"categoryNo":119,"categoryName":120,"taxonomyType":121,"taxonomyTypeDesc":122},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",[124,125,126,127,128,129,130],"methodology","xiaolai","xiaolai methodology","朋友","投资","writing","收入",{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":145,"publishedAt":36},"01M70P13NB9YP7ACJR1S861CDS","zread","Produce and consume a wiki-style knowledge base for a code repository via the `zread` CLI and its on-disk output under `.\u002F.zread\u002Fwiki\u002F`. Use this skill whenever the user wants to understand, onboard onto, explore, summarize, map, or get an overview of an unfamiliar codebase; asks for architecture docs, a project wiki, a repo walkthrough, module\u002Fpackage explanations, or \"what does this repo do\"; wants to generate, regenerate, resume, browse, or serve code documentation locally; or mentions zread \u002F zread.ai directly. Also use it proactively before diving into a large unknown repo — if `.\u002F.zread\u002Fwiki\u002Fcurrent` exists, read the generated pages instead of crawling source file-by-file; if it doesn't, consider offering to run `zread generate`. The trigger is the intent (understand a codebase through generated docs), not the literal word \"zread\".","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fzread\u002FSKILL.md",[137,140],{"categoryNo":138,"categoryName":139,"taxonomyType":121,"taxonomyTypeDesc":122},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":141,"categoryName":142,"taxonomyType":143,"taxonomyTypeDesc":144},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[133,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164],"knowledge","wiki","explore","walkthrough","project","get","knowledge base","run","generate","intent","cli","map","documentation","project wiki","disk","architecture","code","unknown","generated",{"skillNo":166,"name":167,"nameZh":6,"description":168,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":169,"categories":170,"keywords":175,"publishedAt":36},"01M70P10S4SWMZQ8BJAFBFDRW3","youtube","Launch the YouTube Creator Agent to auto-generate videos.\nUse when: user types \u002Fyoutube, wants to create YouTube videos,\ngenerate video content, or upload to YouTube.\nKeywords: youtube, video, 视频, 创作视频, 发布视频","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube\u002FSKILL.md",[171,172],{"categoryNo":138,"categoryName":139,"taxonomyType":121,"taxonomyTypeDesc":122},{"categoryNo":173,"categoryName":174,"taxonomyType":143,"taxonomyTypeDesc":144},"01JBSEEDCAT000000000000036","内容创作者",[167,21,176,177,178,179,180,181,154,182,183,184,26,28,185,186,187,188],"content","视频","creator","generate video","videos","launch","video content","create","upload","auto","types","Video","创作",{"skillNo":190,"name":191,"nameZh":6,"description":192,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":193,"categories":194,"keywords":196,"publishedAt":36},"01M70P0YE95D8BACZCGV8BK90K","youtube-script-sync","Sync, update, or optimize Tony's GetNote knowledge base `Youtube视频逐字稿` by turning each parent recording or first-person idea note into an idempotent child note containing a directly recordable YouTube long-video or Shorts script. Use when the user says “更新\u002F同步\u002F优化 Youtube视频逐字稿知识库”, or says “更新\u002F同步\u002F优化该知识库” while this knowledge base is the active context.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-script-sync\u002FSKILL.md",[195],{"categoryNo":138,"categoryName":139,"taxonomyType":121,"taxonomyTypeDesc":122},[197,198,167,199,200,187,201,202,203,204,205,206,207,208,152,209,26,28,146,210],"sync","script","youtube script sync","recording","shorts","first","base","person","update","active","child","optimize","idempotent","long",{"skillNo":212,"name":213,"nameZh":6,"description":214,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":215,"categories":216,"keywords":221,"publishedAt":36},"01M70P0WMV5XT0EEC0N445PW07","writing-skills","How to write SKILL.md files that trigger reliably and teach effectively. Use when creating, improving, or reviewing Claude Code skills.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fxiaolai\u002Fnlpm\u002Fskills\u002Fnlpm\u002Fwriting-skills\u002FSKILL.md",[217,218,219,220],{"categoryNo":119,"categoryName":120,"taxonomyType":121,"taxonomyTypeDesc":122},{"categoryNo":138,"categoryName":139,"taxonomyType":121,"taxonomyTypeDesc":122},{"categoryNo":141,"categoryName":142,"taxonomyType":143,"taxonomyTypeDesc":144},{"categoryNo":173,"categoryName":174,"taxonomyType":143,"taxonomyTypeDesc":144},[222,129,223,224,225,226,162,26,28,227,228,229,230],"writing skills","how","teach","write skill","write","claude code skills","files","how to","trigger",{"skillNo":232,"name":233,"nameZh":6,"description":234,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":235,"categories":236,"keywords":239,"publishedAt":36},"01M70P0SJFR4YVJQBGMHKGSKXT","writing-rules","How to write .claude\u002Frules\u002F files that Claude actually follows. Use when creating, improving, or reviewing project rules.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fxiaolai\u002Fnlpm\u002Fskills\u002Fnlpm\u002Fwriting-rules\u002FSKILL.md",[237,238],{"categoryNo":119,"categoryName":120,"taxonomyType":121,"taxonomyTypeDesc":122},{"categoryNo":173,"categoryName":174,"taxonomyType":143,"taxonomyTypeDesc":144},[240,129,241,150,242,228,26,223,226,28,229],"Rules","writing rules","project rules"]