[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79S6HFCMPDHD6A02PQGKZFQ":3,"skill-same:01M79S6HFCMPDHD6A02PQGKZFQ":108},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":58,"commitSha":59,"skillDirectory":60,"forksCount":61,"repositoryHtmlUrl":62,"repositoryHomepageUrl":6,"ownerHtmlUrl":63,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":64,"install":65,"resources":77},"01M79S6HFCMPDHD6A02PQGKZFQ","search-orchestrator",null,"搜索编排底座 · 把\"江西附近游玩\"这类模糊地理意图转成结构化查询矩阵，调度online-search \u002F 美团连接器 \u002F xhs-explore skill 做 fan-out → fan-in 多源混合检索，输出带置信度的候选地点 + 内容证据池。是 03\u002F04\u002F06\u002F09 的共用底座。","infometa\u002Fworkbuddyskills","infometa","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F109397516?v=4","01M79FS3XQK361MJFWQ2GFK95C","01M79FS3YPWKZYW9EPAAP2BFBJ",447,"MIT","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: search-orchestrator\ndescription: 搜索编排底座 · 把\"江西附近游玩\"这类模糊地理意图转成结构化查询矩阵，调度online-search \u002F 美团连接器 \u002F xhs-explore skill 做 fan-out → fan-in 多源混合检索，输出带置信度的候选地点 + 内容证据池。是 03\u002F04\u002F06\u002F09 的共用底座。\nversion: 1.0.0\nauthor:\ntags: [travel, search, geo-fanout, multi-source, orchestrator]\nlicense: MIT\ntriggers:\n  - 上游 skill 需要基于地理位置做\"附近 \u002F 周边 \u002F 区域\"模糊搜索时\n  - 用户输入\"X 附近游玩\"、\"X 周边自驾\"、\"X 一日游\"等模糊地理意图\n  - 03 \u002F 04 \u002F 06 \u002F 09 任一 skill 调用前\ninputs:\n  - name: intent\n    type: string\n    doc: 用户原始自然语言意图，如\"江西附近游玩 4 天\"\n    required: true\n  - name: user_profile\n    type: file\n    formats: [json]\n    doc: 来自 preference-load 的偏好画像\n    required: true\n  - name: domain\n    type: string\n    enum: [destination, poi, accommodation, risk]\n    doc: 调用方告诉我搜什么类型，决定查询模板和数据源组合\n    required: true\noutputs:\n  - name: candidates\n    type: file\n    format: json\n    doc: 候选地点 + 每地点的内容证据池（笔记、评论、坐标）+ 数据置信度标记\n---\n\n# 搜索编排底座（search-orchestrator）\n\n## 我解决什么问题\n\n通用 AI 看到\"江西附近游玩\"会笼统说\"庐山很好\"。\n我会做这套事：\n\n1. **地理 fan-out**：调online-search（地理搜索） 拿候选地点池（庐山\u002F婺源\u002F三清山\u002F龙虎山\u002F井冈山\u002F鄱阳湖…）\n2. **查询矩阵生成**：对每个候选地点 × 用户偏好 × 季节 生成 15-20 条精准查询\n3. **多源串行执行**：美团连接器 + xhs-explore skill（含限速\u002F熔断）+ WebSearch 兜底\n4. **去重 + 排序 + 深读**：Top 30 笔记拿全文，Top 5 拿评论\n5. **置信度标记**：每条数据都带 🟢🟡⚪\n\n## 工作流程\n\n```\nintent + user_profile + domain\n   ↓\nStep 1: parse_intent.py        解析中心点 \u002F 半径 \u002F 时间\n   ↓\nStep 2: geo_fanout.py          online-search（地理搜索） → 候选地点 Top 6-10\n   ↓\nStep 3: query_expand.py        候选 × 偏好 × 季节 → 15-20 条查询\n   ↓\nStep 4: 并发执行（≤2 worker，带熔断）\n        ├─ meituan_search.py   调美团连接器\n        ├─ xhs_batch_search.py 调 xhs-explore skill\n        └─ web_fallback.py     WebSearch 兜底\n   ↓\nStep 5: dedupe_rank.py         按 note_id\u002FPOI 去重，按 likes×comments 排序\n   ↓\nStep 6: deep_read.py           Top 30 read 全文 + Top 5 comments\n   ↓\ncandidates.json\n   ↓ 返回上游 skill\n```\n\n## 查询矩阵生成规则（C: 模板 + LLM 混合）\n\n90% 走 `query_templates.json` 的模板填空：\n```json\n{\n  \"destination\": [\n    \"{location} 攻略\",\n    \"{location} 必去\",\n    \"{location} 避雷\",\n    \"{location} {season} 推荐\",\n    \"{location} {duration}日游\"\n  ],\n  \"poi\": [\n    \"{location} {category}\",\n    \"{location} 本地人 {category}\",\n    \"{location} {category} 性价比\"\n  ],\n  \"accommodation\": [\n    \"{location} {tier_label} 酒店推荐\",\n    \"{location} 民宿 真实体验\",\n    \"{location} 住哪里 方便\"\n  ],\n  \"risk\": [\n    \"{location} 避雷\",\n    \"{location} 被坑\",\n    \"{location} 注意事项\"\n  ]\n}\n```\n\n10% LLM 创造性扩展：基于 user_profile 的特殊偏好生成 1-2 条规则覆盖不到的查询，\n如「江西小众老建筑」「成都 不爬山 拍照机位」。\n\n## 反风控约束（与 xhs-explore skill 协同）\n\n- 并发上限 2，超过会触发 461\u002F471\n- 连续 3 次错误 → 立即降级到 WebSearch\n- 同目的地 30 天内查询本地缓存（`data\u002F.cache\u002Fxhs\u002F`）\n\n## 数据置信度\n\n每条返回的 candidate 必须带：\n```json\n{\n  \"name\": \"庐山\",\n  \"source\": \"online-search+xhs\",\n  \"confidence\": \"green\",\n  \"search_data\": { ... },\n  \"xhs_evidence\": [ {note_id, snippet, sentiment} ],\n  \"queried_at\": \"2026-06-03T17:00:00+08:00\"\n}\n```\n\n`confidence` 取值：\n- 🟢 green：MCP\u002FCLI 一手数据\n- 🟡 yellow：WebSearch 抓取\n- ⚪ gray：references\u002F 静态兜底\n\n## 上游 skill 怎么调我\n\n```python\n# 例：destination-research 调用本 skill\nimport subprocess, json\nresult = subprocess.run([\n    \"python\", \"skills\u002Fsearch-orchestrator\u002Fscripts\u002Forchestrate.py\",\n    \"--intent\", \"江西附近游玩 4 天\",\n    \"--profile\", \"data\u002Fuser_profile.json\",\n    \"--domain\", \"destination\",\n    \"--output\", \"data\u002F.cache\u002Frun-{run_id}\u002Fcandidates.json\"\n], capture_output=True, text=True)\ncandidates = json.loads(open(\"data\u002F.cache\u002Frun-xxx\u002Fcandidates.json\").read())\n```\n\n## 反模式\n\n- ❌ 直接调用各种 MCP \u002F CLI，绕开本 skill（会导致重复发请求 \u002F 没有去重 \u002F 没有熔断）\n- ❌ 在上游 skill 自己做查询扩展（应该统一在这里做，保证策略一致）\n- ❌ 不带 domain 参数（不知道用哪套查询模板）\n\n---\n\n_这个 skill 是整个 agent 的\"检索操作系统\"，所有需要外部数据的 skill 都要走这里。_\n","# 搜索编排底座（search-orchestrator）\n\n## 我解决什么问题\n\n通用 AI 看到\"江西附近游玩\"会笼统说\"庐山很好\"。\n我会做这套事：\n\n1. **地理 fan-out**：调online-search（地理搜索） 拿候选地点池（庐山\u002F婺源\u002F三清山\u002F龙虎山\u002F井冈山\u002F鄱阳湖…）\n2. **查询矩阵生成**：对每个候选地点 × 用户偏好 × 季节 生成 15-20 条精准查询\n3. **多源串行执行**：美团连接器 + xhs-explore skill（含限速\u002F熔断）+ WebSearch 兜底\n4. **去重 + 排序 + 深读**：Top 30 笔记拿全文，Top 5 拿评论\n5. **置信度标记**：每条数据都带 🟢🟡⚪\n\n## 工作流程\n\n```\nintent + user_profile + domain\n   ↓\nStep 1: parse_intent.py        解析中心点 \u002F 半径 \u002F 时间\n   ↓\nStep 2: geo_fanout.py          online-search（地理搜索） → 候选地点 Top 6-10\n   ↓\nStep 3: query_expand.py        候选 × 偏好 × 季节 → 15-20 条查询\n   ↓\nStep 4: 并发执行（≤2 worker，带熔断）\n        ├─ meituan_search.py   调美团连接器\n        ├─ xhs_batch_search.py 调 xhs-explore skill\n        └─ web_fallback.py     WebSearch 兜底\n   ↓\nStep 5: dedupe_rank.py         按 note_id\u002FPOI 去重，按 likes×comments 排序\n   ↓\nStep 6: deep_read.py           Top 30 read 全文 + Top 5 comments\n   ↓\ncandidates.json\n   ↓ 返回上游 skill\n```\n\n## 查询矩阵生成规则（C: 模板 + LLM 混合）\n\n90% 走 `query_templates.json` 的模板填空：\n```json\n{\n  \"destination\": [\n    \"{location} 攻略\",\n    \"{location} 必去\",\n    \"{location} 避雷\",\n    \"{location} {season} 推荐\",\n    \"{location} {duration}日游\"\n  ],\n  \"poi\": [\n    \"{location} {category}\",\n    \"{location} 本地人 {category}\",\n    \"{location} {category} 性价比\"\n  ],\n  \"accommodation\": [\n    \"{location} {tier_label} 酒店推荐\",\n    \"{location} 民宿 真实体验\",\n    \"{location} 住哪里 方便\"\n  ],\n  \"risk\": [\n    \"{location} 避雷\",\n    \"{location} 被坑\",\n    \"{location} 注意事项\"\n  ]\n}\n```\n\n10% LLM 创造性扩展：基于 user_profile 的特殊偏好生成 1-2 条规则覆盖不到的查询，\n如「江西小众老建筑」「成都 不爬山 拍照机位」。\n\n## 反风控约束（与 xhs-explore skill 协同）\n\n- 并发上限 2，超过会触发 461\u002F471\n- 连续 3 次错误 → 立即降级到 WebSearch\n- 同目的地 30 天内查询本地缓存（`data\u002F.cache\u002Fxhs\u002F`）\n\n## 数据置信度\n\n每条返回的 candidate 必须带：\n```json\n{\n  \"name\": \"庐山\",\n  \"source\": \"online-search+xhs\",\n  \"confidence\": \"green\",\n  \"search_data\": { ... },\n  \"xhs_evidence\": [ {note_id, snippet, sentiment} ],\n  \"queried_at\": \"2026-06-03T17:00:00+08:00\"\n}\n```\n\n`confidence` 取值：\n- 🟢 green：MCP\u002FCLI 一手数据\n- 🟡 yellow：WebSearch 抓取\n- ⚪ gray：references\u002F 静态兜底\n\n## 上游 skill 怎么调我\n\n```python\n# 例：destination-research 调用本 skill\nimport subprocess, json\nresult = subprocess.run([\n    \"python\", \"skills\u002Fsearch-orchestrator\u002Fscripts\u002Forchestrate.py\",\n    \"--intent\", \"江西附近游玩 4 天\",\n    \"--profile\", \"data\u002Fuser_profile.json\",\n    \"--domain\", \"destination\",\n    \"--output\", \"data\u002F.cache\u002Frun-{run_id}\u002Fcandidates.json\"\n], capture_output=True, text=True)\ncandidates = json.loads(open(\"data\u002F.cache\u002Frun-xxx\u002Fcandidates.json\").read())\n```\n\n## 反模式\n\n- ❌ 直接调用各种 MCP \u002F CLI，绕开本 skill（会导致重复发请求 \u002F 没有去重 \u002F 没有熔断）\n- ❌ 在上游 skill 自己做查询扩展（应该统一在这里做，保证策略一致）\n- ❌ 不带 domain 参数（不知道用哪套查询模板）\n\n---\n\n_这个 skill 是整个 agent 的\"检索操作系统\"，所有需要外部数据的 skill 都要走这里。_\n",{"name":5,"tags":23,"author":6,"inputs":29,"license":14,"outputs":49,"version":53,"triggers":54,"description":7},[24,25,26,27,28],"travel","search","geo-fanout","multi-source","orchestrator",[30,35,41],{"doc":31,"name":32,"type":33,"required":34},"用户原始自然语言意图，如\"江西附近游玩 4 天\"","intent","string",true,{"doc":36,"name":37,"type":38,"formats":39,"required":34},"来自 preference-load 的偏好画像","user_profile","file",[40],"json",{"doc":42,"enum":43,"name":48,"type":33,"required":34},"调用方告诉我搜什么类型，决定查询模板和数据源组合",[44,45,46,47],"destination","poi","accommodation","risk","domain",[50],{"doc":51,"name":52,"type":38,"format":40},"候选地点 + 每地点的内容证据池（笔记、评论、坐标）+ 数据置信度标记","candidates","1.0.0",[55,56,57],"上游 skill 需要基于地理位置做\"附近 \u002F 周边 \u002F 区域\"模糊搜索时","用户输入\"X 附近游玩\"、\"X 周边自驾\"、\"X 一日游\"等模糊地理意图","03 \u002F 04 \u002F 06 \u002F 09 任一 skill 调用前","main","33b6151ccee9aa0816c6f32d021be77d82772755","experts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator",138,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills","https:\u002F\u002Fgithub.com\u002Finfometa","2026-11-11 03:15:35",{"source":8,"skillName":5,"commands":66,"prompt":76,"zipAvailable":34},[67,70,73],{"tool":68,"command":69},"npx","npx skills add infometa\u002Fworkbuddyskills --skill search-orchestrator",{"tool":71,"command":72},"bunx","bunx skills add infometa\u002Fworkbuddyskills --skill search-orchestrator",{"tool":74,"command":75},"pnpm","pnpm dlx skills add infometa\u002Fworkbuddyskills --skill search-orchestrator","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002FSKILL.md",[78,84,88,92,96,100,104],{"path":79,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":82,"sourceHtmlUrl":83},"scripts\u002Fdedupe_rank.py","SCRIPT","脚本",2069,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fdedupe_rank.py",{"path":85,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":86,"sourceHtmlUrl":87},"scripts\u002Fdeep_read.py",2326,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fdeep_read.py",{"path":89,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":90,"sourceHtmlUrl":91},"scripts\u002Fgeo_fanout.py",4414,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fgeo_fanout.py",{"path":93,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":94,"sourceHtmlUrl":95},"scripts\u002Forchestrate.py",6837,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Forchestrate.py",{"path":97,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":98,"sourceHtmlUrl":99},"scripts\u002Fquery_expand.py",3888,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fquery_expand.py",{"path":101,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":102,"sourceHtmlUrl":103},"scripts\u002Fquery_templates.json",1794,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fquery_templates.json",{"path":105,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":106,"sourceHtmlUrl":107},"scripts\u002Fxhs_batch_search.py",5350,"https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fexperts\u002Flazy-travel-planner\u002Fskills\u002Fsearch-orchestrator\u002Fscripts\u002Fxhs_batch_search.py",[109,116,123,130,137,146],{"skillNo":110,"name":111,"nameZh":6,"description":112,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":115,"publishedAt":18},"01M7A1EC013S1PBBTSH2PA2S7D","yingmi-skill","当用户需要查询基金、策略、公告、财经资讯，做资产配置、组合诊断、风险回测、现金流分析，或生成图表、PDF 时，优先使用本 Skill 获取真实数据与可执行能力。","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fyingmi-skill\u002FSKILL.md",[],[],{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":122,"publishedAt":18},"01M7A194T67P6NRVYK8KCTH3CH","web-access","所有联网操作必须通过此 skill 处理，包括：搜索、网页抓取、登录后操作、网络交互等。","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fweb-access\u002FSKILL.md",[],[],{"skillNo":124,"name":125,"nameZh":6,"description":126,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":127,"categories":128,"keywords":129,"publishedAt":18},"01M7A14FATMDD6FK8QAGXY5GVR","tencentmap-map-assistant","腾讯位置服务·地图助手 Skill，一句自然语言调用腾讯地图全套能力，无需开发者账号、开箱即用。提供 AI 旅游攻略、地点搜索（含评分\u002F人均\u002F营业时间）、关键词提示、路线规划（驾车\u002F步行\u002F公交\u002F骑行）、地址解析与逆解析、行政区划、IP 定位、距离计算、天气查询，并可将行程或多 POI 渲染成网页地图。涉及找地点、规划路线、旅游行程、查天气、坐标转换等出行场景时使用。","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Ftencentmap-map-assistant\u002FSKILL.md",[],[],{"skillNo":131,"name":132,"nameZh":6,"description":133,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":134,"categories":135,"keywords":136,"publishedAt":18},"01M7A0SX4XW5053MZ2S921EDH2","skill-scanner","Scan any agent skill for security risks before you install or use it. Powered by Tencent Zhuque Lab A.I.G (AI-Infra-Guard). 100% local static analysis — no file contents or credentials leave your device. Compatible with CodeBuddy, Cursor, Windsurf, Claude Code, OpenClaw and more. Triggers on: 这个 skill 安全吗, skill 安全扫描, 检查 skill 安全, audit skill, scan skill, check skill safety, analyze skill, inspect skill, verify skill, skill security, skill supply chain. Do NOT trigger for general agent usage, full system health checks, project debugging, or normal development.","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fskill-scanner\u002FSKILL.md",[],[],{"skillNo":138,"name":139,"nameZh":6,"description":140,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":141,"licenseFileUrl":142,"sourceHtmlUrl":143,"categories":144,"keywords":145,"publishedAt":18},"01M7A0SQC2CZB3450G8471673M","skill-creator","Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends agent capabilities with specialized knowledge, workflows, or tool integrations.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fskill-creator\u002FSKILL.md",[],[],{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":152,"publishedAt":18},"01M7A0RHV7BKS7DFQF9X9QCCNN","shader-dev","Comprehensive GLSL shader techniques for creating stunning visual effects — ray marching, SDF modeling, fluid simulation, particle systems, procedural generation, lighting, post-processing, and more.","https:\u002F\u002Fgithub.com\u002Finfometa\u002Fworkbuddyskills\u002Fblob\u002Fmain\u002Fskills\u002Fshader-dev\u002FSKILL.md",[],[]]