[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75MDYWDZ2Y9FFVTFFXMZ6DR":3,"skill-same:01M75MDYWDZ2Y9FFVTFFXMZ6DR":60},{"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":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M75MDYWDZ2Y9FFVTFFXMZ6DR","skill-builder",null,"Complete guide for designing, writing, testing, iterating, and packaging Claude Skills (SKILL.md files). Use this skill whenever someone wants to: create a new skill from scratch, improve or update an existing skill, understand the skill format, write a SKILL.md, build a skill folder with scripts and references, evaluate whether a skill is well-written, package a skill for distribution, optimize a skill's trigger description, or understand best practices for Agent Skills. Triggers include phrases like \"build a skill\", \"write a SKILL.md\", \"make a skill for\", \"improve this skill\", \"skill format\", \"create an agent skill\", \"how do I write a skill\", or any request to teach Claude a domain-specific capability through the skills system.","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\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: skill-builder\ndescription: |\n  Complete guide for designing, writing, testing, iterating, and packaging Claude Skills (SKILL.md files). Use this skill whenever someone wants to: create a new skill from scratch, improve or update an existing skill, understand the skill format, write a SKILL.md, build a skill folder with scripts and references, evaluate whether a skill is well-written, package a skill for distribution, optimize a skill's trigger description, or understand best practices for Agent Skills. Triggers include phrases like \"build a skill\", \"write a SKILL.md\", \"make a skill for\", \"improve this skill\", \"skill format\", \"create an agent skill\", \"how do I write a skill\", or any request to teach Claude a domain-specific capability through the skills system.\n---\n\n# Claude Skill Builder\n\nSkills are the primary way to give Claude persistent, domain-specific expertise. A skill is a folder containing a `SKILL.md` file plus optional scripts, references, and assets — all discoverable and loadable on demand without bloating context.\n\n> **Deep dive references:**\n> - `references\u002Fanatomy.md` — structure, file layout, frontmatter spec, progressive disclosure\n> - `references\u002Fwriting.md` — writing patterns, style guide, description engineering, pitfalls\n> - `references\u002Fevals.md` — testing, grading, iteration loop, packaging\n\n---\n\n## The Mental Model: Onboarding Guide for an Expert Hire\n\nBuilding a skill is like writing an onboarding guide for a new hire who is already highly capable but lacks your specific organizational knowledge. They don't need to be told what code is. They need to know: *your* naming conventions, *your* deployment dance, *which* script to run when, and *why* you do things a certain way.\n\nThe implication: **don't explain what Claude already knows**. Only encode knowledge Claude can't infer from its training — your procedures, your environment's quirks, your tools' gotchas, your output standards.\n\n---\n\n## Quick Decision: What Kind of Skill?\n\n| Need | Skill Type |\n|------|-----------|\n| Teach a workflow with multiple steps | **Procedural** — numbered steps, decision tree |\n| Wrap an environment's constraints (libraries, paths, formats) | **Environmental** — code snippets, gotcha table |\n| Layer expert guidance on top of an MCP server | **MCP Enhancement** — workflow + tool guidance |\n| Router that dispatches to other skills\u002Ffiles | **Router** — dispatch table, `references\u002F` per domain |\n| Style or quality standard | **Standard** — constraints, examples of good\u002Fbad |\n\nMost skills combine types. Decide the primary type first; it governs the opening structure.\n\n---\n\n## The Core Loop\n\n```\nIdentify gap → Draft SKILL.md → Test on 2-3 real prompts → Review outputs → \nRevise based on what Claude actually did wrong → Repeat → Package\n```\n\nThis is not waterfall. The most common mistake is writing 500 lines before running a single test. **Start thin, iterate from real failures.**\n\n### Step 1 — Capture Intent Before Writing\n\nAnswer these before touching a file:\n\n1. What will Claude be able to do that it couldn't before?\n2. What does the user say when they need this? (triggers)\n3. What is the output format — a file, a code block, a decision, prose?\n4. What does Claude do *wrong* without the skill? (This is your assertion list.)\n\nIf you can't answer all four, talk to the user more. The answers determine what goes in the skill.\n\n### Step 2 — Write the SKILL.md\n\nStart with the YAML frontmatter, then the body. See `references\u002Fanatomy.md` for the exact format.\n\n**Frontmatter minimum:**\n```yaml\n---\nname: my-skill           # lowercase, hyphens, max 64 chars\ndescription: |\n  What it does and when to trigger it. 1-4 sentences.\n  Be specific about trigger phrases. \"Pushy\" is right — Claude undertriggers.\n---\n```\n\n**Body structure for most skills:**\n```\n# [Skill Name]\n\nBrief framing paragraph (what this is, the mental model).\n\n## Decision Tree \u002F When to Use What\nShort table or if-then. Helps Claude pick the right path fast.\n\n## Core Workflow\nSteps or sections. Explain the *why* behind non-obvious choices.\n\n## Gotchas \u002F Common Mistakes\nWhat Claude will get wrong without this section.\n\n## Quick Reference\nCopy-paste templates, code snippets, lookup tables.\n```\n\nKeep SKILL.md under 500 lines. When you exceed that: extract to `references\u002F`.\n\n### Step 3 — Test on Real Prompts\n\nRun 2-3 prompts that represent what real users would actually say. Observe:\n- Did Claude read the skill? (Check for `bash SKILL.md` in transcript)\n- Did Claude follow the workflow or improvise its own?\n- Did the output match what you'd consider correct?\n- Where did it go off-track?\n\nWrite down *specific* failures — not \"it was bad\" but \"it used pip install without `--break-system-packages`\" or \"it didn't read `references\u002Fforms.md` before filling the form.\"\n\n### Step 4 — Revise Based on Failures\n\nMap each failure to a skill fix:\n\n| Failure Type | Fix |\n|-------------|-----|\n| Claude ignored a step | Make the step more explicit + explain *why* it matters |\n| Claude reinvented a helper script | Bundle the script in `scripts\u002F`, reference it explicitly |\n| Claude read SKILL.md but missed the reference file | Make the reference call-to-action more direct |\n| Output format was wrong | Add a template or example in the skill |\n| Claude undertriggered (didn't read skill when it should) | Strengthen the description; add more trigger phrases |\n| Claude overtriggered (read skill unnecessarily) | Tighten the description with explicit NOT triggers |\n\n### Step 5 — Package\n\nWhen the skill is stable:\n```bash\npython package_skill.py path\u002Fto\u002Fskill-folder\u002F\n```\nThis creates a `.skill` file ready for installation. See `references\u002Fevals.md` for the full packaging checklist.\n\n---\n\n## The Three Laws of Skill Writing\n\n**1. Only say what Claude doesn't already know.**  \nEvery line competes for context window. If Claude could infer it from training, cut it. Domain-specific gotchas, environment paths, output templates — keep. Generic \"be helpful and clear\" — cut.\n\n**2. Explain the why, not just the what.**  \nInstructions without reasons get ignored when Claude thinks it knows better. \"Use `--break-system-packages`\" is weaker than \"Use `--break-system-packages` because pip in this environment rejects installs without it, even for packages you need.\"\n\n**3. Trust progressive disclosure — don't front-load everything.**  \nThe SKILL.md body is Level 2. Reference files are Level 3. Scripts are Level 4 (execute without loading). If information is only needed for a sub-case, move it to a reference file and point to it with a clear trigger.\n\n---\n\n## Common Pitfalls\n\n| Mistake | What Happens | Fix |\n|---------|-------------|-----|\n| Front-loading the full reference manual | SKILL.md hits 800 lines, Claude struggles to extract the key steps | Split into references\u002F — put procedures in SKILL.md, details in references\u002F |\n| Writing ALWAYS\u002FNEVER without explaining why | Claude ignores caps-lock rules when its own judgment says otherwise | Replace with reasoning: \"Do X because Y will happen otherwise\" |\n| Generic description that doesn't name triggers | Skill never fires even when it should | Name the exact phrases users say (\"write a skill\", \"build a SKILL.md\") |\n| Asserting things Claude already knows | Wasted tokens, no improvement | Audit every line: \"Would Claude do this without being told?\" If yes, cut |\n| No evals before shipping | Skill seems fine but fails on real diverse inputs | Always run 3 real prompts before packaging |\n| Bundle everything in one file | Any context means all context — bloated for simple cases | Use progressive disclosure, reference files for optional content |\n\n---\n\n## Quick Reference: File Layout\n\n```\nmy-skill\u002F\n├── SKILL.md           ← Required. Frontmatter + core instructions.\n├── references\u002F        ← Optional. Domain docs, sub-guides.\n│   ├── guide-a.md\n│   └── guide-b.md\n├── scripts\u002F           ← Optional. Pre-built Python\u002Fbash helpers.\n│   └── process.py\n├── assets\u002F            ← Optional. Templates, fonts, icons.\n└── evals\u002F             ← Recommended. Test prompts for iteration.\n    └── evals.json\n```\n\nThe `.skill` file is just a zip of this folder (evals\u002F excluded from distribution).","# Claude Skill Builder\n\nSkills are the primary way to give Claude persistent, domain-specific expertise. A skill is a folder containing a `SKILL.md` file plus optional scripts, references, and assets — all discoverable and loadable on demand without bloating context.\n\n> **Deep dive references:**\n> - `references\u002Fanatomy.md` — structure, file layout, frontmatter spec, progressive disclosure\n> - `references\u002Fwriting.md` — writing patterns, style guide, description engineering, pitfalls\n> - `references\u002Fevals.md` — testing, grading, iteration loop, packaging\n\n---\n\n## The Mental Model: Onboarding Guide for an Expert Hire\n\nBuilding a skill is like writing an onboarding guide for a new hire who is already highly capable but lacks your specific organizational knowledge. They don't need to be told what code is. They need to know: *your* naming conventions, *your* deployment dance, *which* script to run when, and *why* you do things a certain way.\n\nThe implication: **don't explain what Claude already knows**. Only encode knowledge Claude can't infer from its training — your procedures, your environment's quirks, your tools' gotchas, your output standards.\n\n---\n\n## Quick Decision: What Kind of Skill?\n\n| Need | Skill Type |\n|------|-----------|\n| Teach a workflow with multiple steps | **Procedural** — numbered steps, decision tree |\n| Wrap an environment's constraints (libraries, paths, formats) | **Environmental** — code snippets, gotcha table |\n| Layer expert guidance on top of an MCP server | **MCP Enhancement** — workflow + tool guidance |\n| Router that dispatches to other skills\u002Ffiles | **Router** — dispatch table, `references\u002F` per domain |\n| Style or quality standard | **Standard** — constraints, examples of good\u002Fbad |\n\nMost skills combine types. Decide the primary type first; it governs the opening structure.\n\n---\n\n## The Core Loop\n\n```\nIdentify gap → Draft SKILL.md → Test on 2-3 real prompts → Review outputs → \nRevise based on what Claude actually did wrong → Repeat → Package\n```\n\nThis is not waterfall. The most common mistake is writing 500 lines before running a single test. **Start thin, iterate from real failures.**\n\n### Step 1 — Capture Intent Before Writing\n\nAnswer these before touching a file:\n\n1. What will Claude be able to do that it couldn't before?\n2. What does the user say when they need this? (triggers)\n3. What is the output format — a file, a code block, a decision, prose?\n4. What does Claude do *wrong* without the skill? (This is your assertion list.)\n\nIf you can't answer all four, talk to the user more. The answers determine what goes in the skill.\n\n### Step 2 — Write the SKILL.md\n\nStart with the YAML frontmatter, then the body. See `references\u002Fanatomy.md` for the exact format.\n\n**Frontmatter minimum:**\n```yaml\n---\nname: my-skill           # lowercase, hyphens, max 64 chars\ndescription: |\n  What it does and when to trigger it. 1-4 sentences.\n  Be specific about trigger phrases. \"Pushy\" is right — Claude undertriggers.\n---\n```\n\n**Body structure for most skills:**\n```\n# [Skill Name]\n\nBrief framing paragraph (what this is, the mental model).\n\n## Decision Tree \u002F When to Use What\nShort table or if-then. Helps Claude pick the right path fast.\n\n## Core Workflow\nSteps or sections. Explain the *why* behind non-obvious choices.\n\n## Gotchas \u002F Common Mistakes\nWhat Claude will get wrong without this section.\n\n## Quick Reference\nCopy-paste templates, code snippets, lookup tables.\n```\n\nKeep SKILL.md under 500 lines. When you exceed that: extract to `references\u002F`.\n\n### Step 3 — Test on Real Prompts\n\nRun 2-3 prompts that represent what real users would actually say. Observe:\n- Did Claude read the skill? (Check for `bash SKILL.md` in transcript)\n- Did Claude follow the workflow or improvise its own?\n- Did the output match what you'd consider correct?\n- Where did it go off-track?\n\nWrite down *specific* failures — not \"it was bad\" but \"it used pip install without `--break-system-packages`\" or \"it didn't read `references\u002Fforms.md` before filling the form.\"\n\n### Step 4 — Revise Based on Failures\n\nMap each failure to a skill fix:\n\n| Failure Type | Fix |\n|-------------|-----|\n| Claude ignored a step | Make the step more explicit + explain *why* it matters |\n| Claude reinvented a helper script | Bundle the script in `scripts\u002F`, reference it explicitly |\n| Claude read SKILL.md but missed the reference file | Make the reference call-to-action more direct |\n| Output format was wrong | Add a template or example in the skill |\n| Claude undertriggered (didn't read skill when it should) | Strengthen the description; add more trigger phrases |\n| Claude overtriggered (read skill unnecessarily) | Tighten the description with explicit NOT triggers |\n\n### Step 5 — Package\n\nWhen the skill is stable:\n```bash\npython package_skill.py path\u002Fto\u002Fskill-folder\u002F\n```\nThis creates a `.skill` file ready for installation. See `references\u002Fevals.md` for the full packaging checklist.\n\n---\n\n## The Three Laws of Skill Writing\n\n**1. Only say what Claude doesn't already know.**  \nEvery line competes for context window. If Claude could infer it from training, cut it. Domain-specific gotchas, environment paths, output templates — keep. Generic \"be helpful and clear\" — cut.\n\n**2. Explain the why, not just the what.**  \nInstructions without reasons get ignored when Claude thinks it knows better. \"Use `--break-system-packages`\" is weaker than \"Use `--break-system-packages` because pip in this environment rejects installs without it, even for packages you need.\"\n\n**3. Trust progressive disclosure — don't front-load everything.**  \nThe SKILL.md body is Level 2. Reference files are Level 3. Scripts are Level 4 (execute without loading). If information is only needed for a sub-case, move it to a reference file and point to it with a clear trigger.\n\n---\n\n## Common Pitfalls\n\n| Mistake | What Happens | Fix |\n|---------|-------------|-----|\n| Front-loading the full reference manual | SKILL.md hits 800 lines, Claude struggles to extract the key steps | Split into references\u002F — put procedures in SKILL.md, details in references\u002F |\n| Writing ALWAYS\u002FNEVER without explaining why | Claude ignores caps-lock rules when its own judgment says otherwise | Replace with reasoning: \"Do X because Y will happen otherwise\" |\n| Generic description that doesn't name triggers | Skill never fires even when it should | Name the exact phrases users say (\"write a skill\", \"build a SKILL.md\") |\n| Asserting things Claude already knows | Wasted tokens, no improvement | Audit every line: \"Would Claude do this without being told?\" If yes, cut |\n| No evals before shipping | Skill seems fine but fails on real diverse inputs | Always run 3 real prompts before packaging |\n| Bundle everything in one file | Any context means all context — bloated for simple cases | Use progressive disclosure, reference files for optional content |\n\n---\n\n## Quick Reference: File Layout\n\n```\nmy-skill\u002F\n├── SKILL.md           ← Required. Frontmatter + core instructions.\n├── references\u002F        ← Optional. Domain docs, sub-guides.\n│   ├── guide-a.md\n│   └── guide-b.md\n├── scripts\u002F           ← Optional. Pre-built Python\u002Fbash helpers.\n│   └── process.py\n├── assets\u002F            ← Optional. Templates, fonts, icons.\n└── evals\u002F             ← Recommended. Test prompts for iteration.\n    └── evals.json\n```\n\nThe `.skill` file is just a zip of this folder (evals\u002F excluded from distribution).",{"name":5,"description":7},"main","93f2582acd699b8d3917b2b61fdd2178b69f921c","src\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder",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":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add Razaib-khan\u002FForgeWeave --skill skill-builder",{"tool":38,"command":39},"bunx","bunx skills add Razaib-khan\u002FForgeWeave --skill skill-builder",{"tool":41,"command":42},"pnpm","pnpm dlx skills add Razaib-khan\u002FForgeWeave --skill skill-builder","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder\u002FSKILL.md",true,[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"references\u002Fanatomy.md","REFERENCE","参考文档",7449,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder\u002Freferences\u002Fanatomy.md",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fevals.md",9439,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder\u002Freferences\u002Fevals.md",{"path":57,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fwriting.md",10122,"https:\u002F\u002Fgithub.com\u002FRazaib-khan\u002FForgeWeave\u002Fblob\u002Fmain\u002Fsrc\u002Fforgeweave\u002Ftemplates\u002Fqwen\u002F.qwen\u002Fskills\u002Fskill-builder\u002Freferences\u002Fwriting.md",[61,68,75,82,89,96],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"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":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"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":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"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":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"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":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"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",[],[]]