[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75MASWH3K2W82XGE4ES7TAN":3,"skill-same:01M75MASWH3K2W82XGE4ES7TAN":98},{"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":36,"commitSha":37,"skillDirectory":38,"forksCount":39,"repositoryHtmlUrl":40,"repositoryHomepageUrl":6,"ownerHtmlUrl":41,"ownerBlogUrl":6,"ownerDescription":42,"updatedAt":43,"install":44,"resources":57},"01M75MASWH3K2W82XGE4ES7TAN","Deep Research",null,"Multi-stage research pipeline that decomposes a topic, gathers structured information, validates it, and synthesizes a final report. Supports formatted\u002Funformatted output and skill\u002Fno-skill generation modes.","Razaib-khan\u002FForgeWeave","Razaib-khan","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160170334?v=4","01M75M7QPJVP3N29AZVG6RN6R7","Razaib Khan","01M75M7QQS6KEPD0W2ZB99ENKK",1,"MIT","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nskill_id: deep-research\nname: Deep Research\nversion: 1.0.0\ndescription: Multi-stage research pipeline that decomposes a topic, gathers structured information, validates it, and synthesizes a final report. Supports formatted\u002Funformatted output and skill\u002Fno-skill generation modes.\nauthor: ForgeWeave Core\ntui_compatibility:\n  - opencode\n  - claude\n  - gemini\n  - qwen\ntags:\n  - research\n  - pipeline\n  - synthesis\n---\n\n# Deep Research\n\n## Purpose\n\nExecute a multi-stage research pipeline that decomposes a vague topic into structured subtopics, gathers usage-focused information from authoritative sources via parallel agents, validates claims for consistency, produces a synthesis-grade report, and converts raw findings into a reusable industry-grade skill. This is invoked internally by the deep-research skill — never call the pipeline stages directly.\n\n## When to Use\n\n- A comprehensive, multi-faceted report is needed covering 3+ subtopics\n- The topic requires crawling multiple authoritative sources (docs, API refs, guides)\n- The output must be structured, validated, and free of hallucination\n- The question cannot be answered by a single source or quick lookup\n\n## When Not to Use\n\n- A quick factual answer is needed — use websearch instead\n- Only one source needs to be checked — use the MCP data plane tools directly\n- The topic is a simple how-to question — answer directly\n- Real-time data is needed (stock prices, live scores) — use web-research\n\n## Inputs\n\n| Input | Type | Required | Description |\n|---|---|---|---|\n| `topic` | string | Yes | The research topic or question |\n| `depth` | enum | No (default: standard) | quick, standard, deep |\n| `focus` | enum | No (default: usage) | usage, architecture, comparison, general |\n| `constraints` | string | No | Additional rules from AGENTS.md or user |\n| `output_mode` | enum | No (default: formatted) | formatted, unformatted |\n| `skill_mode` | enum | No (default: skill) | skill, no-skill |\n\n## Expected Outputs\n\n| Output | Condition | Description |\n|---|---|---|\n| `research\u002F\u003Cslug>-plan.md` | Always | Structured plan with subtopics, questions, seed URLs |\n| `research\u002F\u003Cslug>-raw\u002F` | Always | One file per subtopic from parallel research agents |\n| `research\u002F\u003Cslug>-validated.md` | Always | Cross-checked, deduplicated, hallucination-filtered |\n| `research\u002F\u003Cslug>-report.md` | `formatted` mode | Final synthesis with all findings, code examples, sources |\n| `.opencode\u002Fskills\u002F\u003Ctopic>\u002FSKILL.md` | `skill` mode | Reusable skill generated from findings |\n\n## Internal Workflow\n\n### Stage 1: Plan\nInternal planner agent decomposes topic into 3-7 subtopics with questions and seed URLs from authoritative sources only. For JS-rendered content, Playwright MCP tools (`browser_navigate`, `browser_snapshot`) are available for interactive browsing during URL discovery.\n\n### Stage 2: Research (Parallel)\nInternal research agents run concurrently — one per subtopic. Each crawls seed URLs, extracts code examples, API signatures, and edge cases. For JS-rendered pages, agents use Playwright MCP tools (`browser_navigate`, `browser_snapshot`).\n\n### Stage 3: Validate\nInternal validator cross-checks all subtopic outputs: removes unsupported claims, flags contradictions, deduplicates findings.\n\n### Stage 4: Synthesize\nInternal synthesizer merges validated research into a final report with sections: Overview, Getting Started, Core Content, Advanced Patterns, Migration Guide, Best Practices, Edge Cases, Sources.\n\n### Stage 5: Output\nInternal output writer saves research results based on output mode:\n- `formatted` (default): Produces a structured report at `research\u002F\u003Cslug>-report.md` with sections: Overview, Getting Started, Core Content, Advanced Patterns, Migration Guide, Best Practices, Edge Cases, Sources\n- `unformatted`: Saves raw scraped data as individual markdown files in `research\u002F\u003Cslug>-raw\u002F` with minimal processing\n\nRaw data is always saved to `research\u002F\u003Cslug>-raw\u002F` regardless of mode.\n\n### Stage 6: Skill Conversion (conditional)\nAI uses the skill-builder skill to convert the raw findings into a reusable SKILL.md file — **only when `skill` mode is selected**:\n- Reads all `research\u002F\u003Cslug>-raw\u002F*.md` files\n- Identifies reusable patterns, APIs, and best practices\n- Writes a structured SKILL.md with frontmatter, workflow steps, gotchas, and references\n- Places the skill in `.opencode\u002Fskills\u002F\u003Ctopic-slug>\u002F` for future coding use\n- Reports the skill path to the user\n\n## Required Checks\n\n- [ ] Planner ran first (never skip)\n- [ ] Research agents ran in parallel\n- [ ] Validator ran after all research completed\n- [ ] Every claim has a source URL\n- [ ] No blog posts or changelogs used as sources\n\n## Failure Modes\n\n| Failure Condition | Response |\n|---|---|\n| Planner produces \u003C3 subtopics | Respawn with \"produce at least 3\" |\n| All seed URLs return 404 | Report \"all sources unavailable\" |\n| Validator flags >50% claims as unsupported | Re-run research with better URLs |\n| Pipeline exceeds max iterations | Stop and return partial results |\n\n## Examples\n\n### Example 1: Full research with skill (default)\nTrigger: `\u002Fforge-research Next.js 16 caching --depth=deep`\nResult: `research\u002Fnextjs16-report.md` + `.opencode\u002Fskills\u002Fnextjs16\u002FSKILL.md`\n\n### Example 2: Raw output only, no skill\nTrigger: `\u002Fforge-research Next.js 16 caching unformatted no-skill`\nResult: `research\u002Fnextjs16-raw\u002F*.md` (raw files only)\n\n### Example 3: Formatted report, no skill\nTrigger: `\u002Fforge-research Python 3.14 pattern matching formed no-skill`\nResult: `research\u002Fpython314-report.md` (structured report only)\n\n## References\n\n| Reference | Path |\n|---|---|\n| RESEARCH_INSTRUCTIONS.md | `.\u002FRESEARCH_INSTRUCTIONS.md` |\n","# Deep Research\n\n## Purpose\n\nExecute a multi-stage research pipeline that decomposes a vague topic into structured subtopics, gathers usage-focused information from authoritative sources via parallel agents, validates claims for consistency, produces a synthesis-grade report, and converts raw findings into a reusable industry-grade skill. This is invoked internally by the deep-research skill — never call the pipeline stages directly.\n\n## When to Use\n\n- A comprehensive, multi-faceted report is needed covering 3+ subtopics\n- The topic requires crawling multiple authoritative sources (docs, API refs, guides)\n- The output must be structured, validated, and free of hallucination\n- The question cannot be answered by a single source or quick lookup\n\n## When Not to Use\n\n- A quick factual answer is needed — use websearch instead\n- Only one source needs to be checked — use the MCP data plane tools directly\n- The topic is a simple how-to question — answer directly\n- Real-time data is needed (stock prices, live scores) — use web-research\n\n## Inputs\n\n| Input | Type | Required | Description |\n|---|---|---|---|\n| `topic` | string | Yes | The research topic or question |\n| `depth` | enum | No (default: standard) | quick, standard, deep |\n| `focus` | enum | No (default: usage) | usage, architecture, comparison, general |\n| `constraints` | string | No | Additional rules from AGENTS.md or user |\n| `output_mode` | enum | No (default: formatted) | formatted, unformatted |\n| `skill_mode` | enum | No (default: skill) | skill, no-skill |\n\n## Expected Outputs\n\n| Output | Condition | Description |\n|---|---|---|\n| `research\u002F\u003Cslug>-plan.md` | Always | Structured plan with subtopics, questions, seed URLs |\n| `research\u002F\u003Cslug>-raw\u002F` | Always | One file per subtopic from parallel research agents |\n| `research\u002F\u003Cslug>-validated.md` | Always | Cross-checked, deduplicated, hallucination-filtered |\n| `research\u002F\u003Cslug>-report.md` | `formatted` mode | Final synthesis with all findings, code examples, sources |\n| `.opencode\u002Fskills\u002F\u003Ctopic>\u002FSKILL.md` | `skill` mode | Reusable skill generated from findings |\n\n## Internal Workflow\n\n### Stage 1: Plan\nInternal planner agent decomposes topic into 3-7 subtopics with questions and seed URLs from authoritative sources only. For JS-rendered content, Playwright MCP tools (`browser_navigate`, `browser_snapshot`) are available for interactive browsing during URL discovery.\n\n### Stage 2: Research (Parallel)\nInternal research agents run concurrently — one per subtopic. Each crawls seed URLs, extracts code examples, API signatures, and edge cases. For JS-rendered pages, agents use Playwright MCP tools (`browser_navigate`, `browser_snapshot`).\n\n### Stage 3: Validate\nInternal validator cross-checks all subtopic outputs: removes unsupported claims, flags contradictions, deduplicates findings.\n\n### Stage 4: Synthesize\nInternal synthesizer merges validated research into a final report with sections: Overview, Getting Started, Core Content, Advanced Patterns, Migration Guide, Best Practices, Edge Cases, Sources.\n\n### Stage 5: Output\nInternal output writer saves research results based on output mode:\n- `formatted` (default): Produces a structured report at `research\u002F\u003Cslug>-report.md` with sections: Overview, Getting Started, Core Content, Advanced Patterns, Migration Guide, Best Practices, Edge Cases, Sources\n- `unformatted`: Saves raw scraped data as individual markdown files in `research\u002F\u003Cslug>-raw\u002F` with minimal processing\n\nRaw data is always saved to `research\u002F\u003Cslug>-raw\u002F` regardless of mode.\n\n### Stage 6: Skill Conversion (conditional)\nAI uses the skill-builder skill to convert the raw findings into a reusable SKILL.md file — **only when `skill` mode is selected**:\n- Reads all `research\u002F\u003Cslug>-raw\u002F*.md` files\n- Identifies reusable patterns, APIs, and best practices\n- Writes a structured SKILL.md with frontmatter, workflow steps, gotchas, and references\n- Places the skill in `.opencode\u002Fskills\u002F\u003Ctopic-slug>\u002F` for future coding use\n- Reports the skill path to the user\n\n## Required Checks\n\n- [ ] Planner ran first (never skip)\n- [ ] Research agents ran in parallel\n- [ ] Validator ran after all research completed\n- [ ] Every claim has a source URL\n- [ ] No blog posts or changelogs used as sources\n\n## Failure Modes\n\n| Failure Condition | Response |\n|---|---|\n| Planner produces \u003C3 subtopics | Respawn with \"produce at least 3\" |\n| All seed URLs return 404 | Report \"all sources unavailable\" |\n| Validator flags >50% claims as unsupported | Re-run research with better URLs |\n| Pipeline exceeds max iterations | Stop and return partial results |\n\n## Examples\n\n### Example 1: Full research with skill (default)\nTrigger: `\u002Fforge-research Next.js 16 caching --depth=deep`\nResult: `research\u002Fnextjs16-report.md` + `.opencode\u002Fskills\u002Fnextjs16\u002FSKILL.md`\n\n### Example 2: Raw output only, no skill\nTrigger: `\u002Fforge-research Next.js 16 caching unformatted no-skill`\nResult: `research\u002Fnextjs16-raw\u002F*.md` (raw files only)\n\n### Example 3: Formatted report, no skill\nTrigger: `\u002Fforge-research Python 3.14 pattern matching formed no-skill`\nResult: `research\u002Fpython314-report.md` (structured report only)\n\n## References\n\n| Reference | Path |\n|---|---|\n| RESEARCH_INSTRUCTIONS.md | `.\u002FRESEARCH_INSTRUCTIONS.md` |\n",{"name":5,"tags":24,"author":28,"version":29,"skill_id":30,"description":7,"tui_compatibility":31},[25,26,27],"research","pipeline","synthesis","ForgeWeave Core","1.0.0","deep-research",[32,33,34,35],"opencode","claude","gemini","qwen","main","93f2582acd699b8d3917b2b61fdd2178b69f921c","src\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research",0,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave","https:\u002F\u002Fgithub.com\u002FRazaib-khan","Web dev & AI explorer | Building, breaking, learning | Long-term mindset","2026-06-22 22:46:11",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add Razaib-khan\u002FForgeWeave --skill \"Deep Research\"",{"tool":50,"command":51},"bunx","bunx skills add Razaib-khan\u002FForgeWeave --skill \"Deep Research\"",{"tool":53,"command":54},"pnpm","pnpm dlx skills add Razaib-khan\u002FForgeWeave --skill \"Deep Research\"","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002FSKILL.md",true,[58,64,68,72,76,82,86,90,94],{"path":59,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":62,"sourceHtmlUrl":63},"evals\u002Ftest_happy_path.json","REFERENCE","参考文档",398,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Fevals\u002Ftest_happy_path.json",{"path":65,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":66,"sourceHtmlUrl":67},"examples\u002FREADME.md",700,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Fexamples\u002FREADME.md",{"path":69,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Fpipeline-architecture.md",2610,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Freferences\u002Fpipeline-architecture.md",{"path":73,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Furl-sourcing-rules.md",1212,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Freferences\u002Furl-sourcing-rules.md",{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"scripts\u002Fplan_generator.py","SCRIPT","脚本",2822,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Fscripts\u002Fplan_generator.py",{"path":83,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":84,"sourceHtmlUrl":85},"scripts\u002Fsynthesize_report.py",2663,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Fscripts\u002Fsynthesize_report.py",{"path":87,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":88,"sourceHtmlUrl":89},"scripts\u002Fvalidate_research.py",2259,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Fscripts\u002Fvalidate_research.py",{"path":91,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":92,"sourceHtmlUrl":93},"templates\u002Fplan_template.md",413,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Ftemplates\u002Fplan_template.md",{"path":95,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":96,"sourceHtmlUrl":97},"templates\u002Freport_template.md",918,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fgemini\u002F.gemini\u002Fskills\u002Fdeep-research\u002Ftemplates\u002Freport_template.md",[99,106,113,120,127,134],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":103,"categories":104,"keywords":105,"publishedAt":19},"01M75MEJ4J1S2ZERWSM5ZBKRTE","Workflow Orchestrator","Coordinates multi-skill execution pipelines (e.g., research → plan → build → test)","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fworkflow-orchestrator\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":19},"01M75MEEZQ0V89BABKZF5XQ3MW","Web Research","Browser-driven or scraping-based skill for real-time internet extraction using Playwright and HTTP-based fetching","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fweb-research\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":19},"01M75MECJV3NEN6GY89ZT7NWFA","Validation Engine","Checks outputs from other skills for consistency, correctness, and adherence to AGENTS.md rules","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fvalidation-engine\u002FSKILL.md",[],[],{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":19},"01M75MEAAXKRA9HJWWQH0MMBGY","Tool Selector","Determines which tools, APIs, MCP actions, or system capabilities are required for a given task","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Ftool-selector\u002FSKILL.md",[],[],{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":19},"01M75ME7JAXQG26T1NEB3VNDM6","Test Generator","Creates test plans and test cases (unit, integration, edge cases) based on existing or planned code","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Ftest-generator\u002FSKILL.md",[],[],{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":140,"publishedAt":19},"01M75ME52949X3D8JQSYAWGDRB","Skill Generator","Meta-skill that creates new SKILL.md files from research outputs and patterns discovered in execution","https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-generator\u002FSKILL.md",[],[]]