[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MWJ1YYR5ZJ5QA4WAYDFNNW":3,"skill-same:01M6MWJ1YYR5ZJ5QA4WAYDFNNW":85},{"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":27,"publishedAt":48,"linkOnly":49,"markdownContent":50,"markdownBody":51,"frontMatter":52,"refName":53,"commitSha":54,"skillDirectory":55,"forksCount":56,"repositoryHtmlUrl":57,"repositoryHomepageUrl":6,"ownerHtmlUrl":58,"ownerBlogUrl":6,"ownerDescription":59,"updatedAt":60,"install":61,"resources":74},"01M6MWJ1YYR5ZJ5QA4WAYDFNNW","agent-docs",null,"Create documentation optimized for AI agent consumption. Use when writing SKILL.md files, README files, API docs, or any documentation that will be read by LLMs in context windows. Helps structure content for RAG retrieval, token efficiency, and the Hybrid Context Hierarchy.","Kernel8901\u002Fai-agent-skills-classification","Kernel8901","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F72647553?v=4","01M6KV388J2H5PNC8D8P940BJD","01M6KV38CS5B3AY63RKZB410YM",7,"MIT","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fagent-docs\u002FSKILL.md",[17,22],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47],"docs","agent docs","hybrid","writing","hierarchy","token efficiency","for","token","efficiency","structure","files","context","read","api","create","rag","documentation","windows","ai agent","llms","2026-11-10 10:12:30",false,"---\nname: agent-docs\ndescription: Create documentation optimized for AI agent consumption. Use when writing SKILL.md files, README files, API docs, or any documentation that will be read by LLMs in context windows. Helps structure content for RAG retrieval, token efficiency, and the Hybrid Context Hierarchy.\n---\n\n# Agent Docs\n\nWrite documentation that AI agents can efficiently consume. Based on Vercel benchmarks and industry standards (AGENTS.md, llms.txt, CLAUDE.md).\n\n## The Hybrid Context Hierarchy\n\nThree-layer architecture for optimal agent performance:\n\n### Layer 1: Constitution (Inline)\n**Always in context.** 2,000–4,000 tokens max.\n\n```markdown\n# AGENTS.md\n> Context: Next.js 16 | Tailwind | Supabase\n\n## ? CRITICAL\n- NO SECRETS in output\n- Use `app\u002F` directory ONLY\n\n## ? DOCS INDEX (use read_file)\n- Auth: `docs\u002Fauth\u002Fllms.txt`\n- DB: `docs\u002Fdb\u002Fschema.md`\n```\n\n**Include:**\n- Security rules, architecture constraints\n- Build\u002Ftest\u002Flint commands (top for primacy bias)\n- Documentation map (where to find more)\n\n### Layer 2: Reference Library (Local Retrieval)\n**Fetched on demand.** 1K–5K token chunks.\n\n- Framework-specific guides\n- Detailed style guides\n- API schemas\n\n### Layer 3: Research Assistant (External)\n**Gated by allow-lists.** Edge cases only.\n\n- Latest library updates\n- Stack Overflow for obscure errors\n- Third-party llms.txt\n\n## Why This Works\n\n**Vercel Benchmark (2026):**\n| Approach | Pass Rate |\n|----------|-----------|\n| Tool-based retrieval | 53% |\n| Retrieval + prompting | 79% |\n| **Inline AGENTS.md** | **100%** |\n\n**Root cause:** Meta-cognitive failure. Agents don't know what they don't know—they assume training data is sufficient. Inline docs bypass this entirely.\n\n## Core Principles\n\n### 1. Compressed Index > Full Docs\n\nAn 8KB compressed index outperforms a 40KB full dump.\n\n**Compress to:**\n- File paths (where code lives)\n- Function signatures (names + types only)\n- Negative constraints (\"Do NOT use X\")\n\n### 2. Structure for Chunking\n\nRAG systems split at headers. Each section must be self-contained:\n\n```markdown\n## Database Setup          ← Chunk boundary\n\nPrerequisites: PostgreSQL 14+\n\n1. Create database...\n```\n\n**Rules:**\n- Front-load key info (chunkers truncate)\n- Descriptive headers (agents search by header text)\n\n### 3. Inline Over Links\n\nAgents can't autonomously browse. Each link = tool call + latency + potential failure.\n\n| Approach | Token Load | Agent Success |\n|----------|------------|---------------|\n| Full inline | ~12K | ✅ High |\n| Links only | ~2K | ❌ Requires fetching |\n| Hybrid | ~4K base | ✅ Best of both |\n\n### 4. The \"Lost in the Middle\" Problem\n\nLLMs have U-shaped attention:\n- **Strong:** Start of context (primacy)\n- **Strong:** End of context (recency)\n- **Weak:** Middle of context\n\n**Solution:** Put critical rules at TOP of AGENTS.md. Governance first, details later.\n\n### 5. Signal-to-Noise Ratio\n\nStrip everything that isn't essential:\n- No \"Welcome to...\" preambles\n- No marketing text\n- No changelogs in core docs\n\nFormats like llms.txt and AGENTS.md mechanically increase SNR.\n\n## llms.txt Standard\n\nMachine-readable doc index for agents:\n\n```markdown\n# Project Name\n\n> One-line project description.\n\n## Authentication\n\n- [Setup](docs\u002Fauth\u002Fsetup.md): Environment vars and init\n- [Server](docs\u002Fauth\u002Fserver.md): Cookie handling\n\n## Database\n\n- [Schema](docs\u002Fdb\u002Fschema.md): Full Prisma schema\n```\n\n**Location:** `\u002Fllms.txt` at domain root\n**Companion:** `\u002Fllms-full.txt` — full concatenated docs, HTML stripped\n\n## Security Considerations\n\n### Inline = Trusted\nAGENTS.md is part of your codebase. Controlled, version-pinned.\n\n### External = Attack Surface\n- Indirect prompt injection via hidden text\n- SSRF risks if agents can browse freely\n- Dependency on external uptime\n\n**Mitigation:** Domain allow-lists, human-in-the-loop for external retrieval.\n\n## Anti-Patterns\n\n1. **Pasting 50 pages** — triggers \"Lost in the Middle\"\n2. **\"See external docs\"** — agents can't browse autonomously\n3. **Generic advice** — \"Write clean code\" (use specific constraints)\n4. **TOC-only docs** — indexes without content\n5. **Trusting retrieval alone** — 53% vs 100% pass rate\n\n## Advanced Patterns\n\nFor detailed guidance on RAG optimization, multi-framework docs, and API templates, see [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\n\n## Validation Checklist\n\n- [ ] Critical governance at TOP of doc\n- [ ] Total inline context under 4K tokens\n- [ ] Each H2 section self-contained\n- [ ] No external links without inline summary\n- [ ] Negative constraints explicit (\"Do NOT...\")\n- [ ] File paths and signatures, not full code\n","# Agent Docs\n\nWrite documentation that AI agents can efficiently consume. Based on Vercel benchmarks and industry standards (AGENTS.md, llms.txt, CLAUDE.md).\n\n## The Hybrid Context Hierarchy\n\nThree-layer architecture for optimal agent performance:\n\n### Layer 1: Constitution (Inline)\n**Always in context.** 2,000–4,000 tokens max.\n\n```markdown\n# AGENTS.md\n> Context: Next.js 16 | Tailwind | Supabase\n\n## ? CRITICAL\n- NO SECRETS in output\n- Use `app\u002F` directory ONLY\n\n## ? DOCS INDEX (use read_file)\n- Auth: `docs\u002Fauth\u002Fllms.txt`\n- DB: `docs\u002Fdb\u002Fschema.md`\n```\n\n**Include:**\n- Security rules, architecture constraints\n- Build\u002Ftest\u002Flint commands (top for primacy bias)\n- Documentation map (where to find more)\n\n### Layer 2: Reference Library (Local Retrieval)\n**Fetched on demand.** 1K–5K token chunks.\n\n- Framework-specific guides\n- Detailed style guides\n- API schemas\n\n### Layer 3: Research Assistant (External)\n**Gated by allow-lists.** Edge cases only.\n\n- Latest library updates\n- Stack Overflow for obscure errors\n- Third-party llms.txt\n\n## Why This Works\n\n**Vercel Benchmark (2026):**\n| Approach | Pass Rate |\n|----------|-----------|\n| Tool-based retrieval | 53% |\n| Retrieval + prompting | 79% |\n| **Inline AGENTS.md** | **100%** |\n\n**Root cause:** Meta-cognitive failure. Agents don't know what they don't know—they assume training data is sufficient. Inline docs bypass this entirely.\n\n## Core Principles\n\n### 1. Compressed Index > Full Docs\n\nAn 8KB compressed index outperforms a 40KB full dump.\n\n**Compress to:**\n- File paths (where code lives)\n- Function signatures (names + types only)\n- Negative constraints (\"Do NOT use X\")\n\n### 2. Structure for Chunking\n\nRAG systems split at headers. Each section must be self-contained:\n\n```markdown\n## Database Setup          ← Chunk boundary\n\nPrerequisites: PostgreSQL 14+\n\n1. Create database...\n```\n\n**Rules:**\n- Front-load key info (chunkers truncate)\n- Descriptive headers (agents search by header text)\n\n### 3. Inline Over Links\n\nAgents can't autonomously browse. Each link = tool call + latency + potential failure.\n\n| Approach | Token Load | Agent Success |\n|----------|------------|---------------|\n| Full inline | ~12K | ✅ High |\n| Links only | ~2K | ❌ Requires fetching |\n| Hybrid | ~4K base | ✅ Best of both |\n\n### 4. The \"Lost in the Middle\" Problem\n\nLLMs have U-shaped attention:\n- **Strong:** Start of context (primacy)\n- **Strong:** End of context (recency)\n- **Weak:** Middle of context\n\n**Solution:** Put critical rules at TOP of AGENTS.md. Governance first, details later.\n\n### 5. Signal-to-Noise Ratio\n\nStrip everything that isn't essential:\n- No \"Welcome to...\" preambles\n- No marketing text\n- No changelogs in core docs\n\nFormats like llms.txt and AGENTS.md mechanically increase SNR.\n\n## llms.txt Standard\n\nMachine-readable doc index for agents:\n\n```markdown\n# Project Name\n\n> One-line project description.\n\n## Authentication\n\n- [Setup](docs\u002Fauth\u002Fsetup.md): Environment vars and init\n- [Server](docs\u002Fauth\u002Fserver.md): Cookie handling\n\n## Database\n\n- [Schema](docs\u002Fdb\u002Fschema.md): Full Prisma schema\n```\n\n**Location:** `\u002Fllms.txt` at domain root\n**Companion:** `\u002Fllms-full.txt` — full concatenated docs, HTML stripped\n\n## Security Considerations\n\n### Inline = Trusted\nAGENTS.md is part of your codebase. Controlled, version-pinned.\n\n### External = Attack Surface\n- Indirect prompt injection via hidden text\n- SSRF risks if agents can browse freely\n- Dependency on external uptime\n\n**Mitigation:** Domain allow-lists, human-in-the-loop for external retrieval.\n\n## Anti-Patterns\n\n1. **Pasting 50 pages** — triggers \"Lost in the Middle\"\n2. **\"See external docs\"** — agents can't browse autonomously\n3. **Generic advice** — \"Write clean code\" (use specific constraints)\n4. **TOC-only docs** — indexes without content\n5. **Trusting retrieval alone** — 53% vs 100% pass rate\n\n## Advanced Patterns\n\nFor detailed guidance on RAG optimization, multi-framework docs, and API templates, see [references\u002Fadvanced-patterns.md](references\u002Fadvanced-patterns.md).\n\n## Validation Checklist\n\n- [ ] Critical governance at TOP of doc\n- [ ] Total inline context under 4K tokens\n- [ ] Each H2 section self-contained\n- [ ] No external links without inline summary\n- [ ] Negative constraints explicit (\"Do NOT...\")\n- [ ] File paths and signatures, not full code\n",{"name":5,"description":7},"main","f4cb2d49f7667e79ef7804f50396d5aaffdf1f56","人工智能与大型语言模型\u002Fagent-docs",1,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification","https:\u002F\u002Fgithub.com\u002FKernel8901","a little little little dobe in life but science geek","2026-04-05 16:43:56",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":73},[63,66,69],{"tool":64,"command":65},"npx","npx skills add Kernel8901\u002Fai-agent-skills-classification --skill agent-docs",{"tool":67,"command":68},"bunx","bunx skills add Kernel8901\u002Fai-agent-skills-classification --skill agent-docs",{"tool":70,"command":71},"pnpm","pnpm dlx skills add Kernel8901\u002Fai-agent-skills-classification --skill agent-docs","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fagent-docs\u002FSKILL.md",true,[75,81],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"_meta.json","REFERENCE","参考文档",276,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fagent-docs\u002F_meta.json",{"path":82,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fadvanced-patterns.md",5402,"https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F人工智能与大型语言模型\u002Fagent-docs\u002Freferences\u002Fadvanced-patterns.md",[86,110,138,149,159,188],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":90,"categories":91,"keywords":92,"publishedAt":48},"01M6Q5R6WZ06RRB9RQTW54B34B","zyfai","Earn yield on any Ethereum wallet on Base, Arbitrum, and Plasma. Use when a user wants passive DeFi yield on their funds. Deploys a non-custodial subaccount (Safe) linked to their EOA, enables automated yield optimization, and lets them deposit\u002Fwithdraw anytime.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzyfai-sdk\u002FSKILL.md",[],[88,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109],"safe","ethereum","defi","Optimization","user","use","Wallet","base","deposit","withdraw","yield","arbitrum","non-custodial","passive","defi yield","earn","yield optimization",{"skillNo":111,"name":112,"nameZh":6,"description":6,"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":122,"publishedAt":48},"01M6Q5R337B5JSNER4P4YS03TN","zerion-api","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api\u002FSKILL.md",[115,118,119],{"categoryNo":116,"categoryName":117,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":120,"categoryName":121,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000030","开发工程师",[41,123,124,125,126,127,128,95,129,99,130,131,132,133,134,135,136,35,137],"zerion api","zerion","MCP","mcp server","server","blockchain","transactions","crypto wallet","crypto","evm","solana","query","across","positions","prices",{"skillNo":139,"name":140,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":141,"categories":142,"keywords":146,"publishedAt":48},"01M6Q5QZFYTDAZMZ27BN5VZXAR","zerion-api-skill","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api-skill\u002FSKILL.md",[143,144,145],{"categoryNo":116,"categoryName":117,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":120,"categoryName":121,"taxonomyType":25,"taxonomyTypeDesc":26},[41,147,148,123,124,125,126,127,128,95,129,99,131,132,133,134,135,136,35,137],"api skill","zerion api skill",{"skillNo":150,"name":151,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":157,"publishedAt":48},"01M6Q5QW17QW98RA3FE45E4RD8","zerion-api-mcp","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fzerion-api-mcp\u002FSKILL.md",[154,155,156],{"categoryNo":116,"categoryName":117,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},{"categoryNo":120,"categoryName":121,"taxonomyType":25,"taxonomyTypeDesc":26},[125,41,158,123,124,126,127,95,129,99,130,131,132,133,134,135,136,35,137],"zerion api mcp",{"skillNo":160,"name":161,"nameZh":6,"description":162,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":163,"categories":164,"keywords":168,"publishedAt":48},"01M6Q5QQYZB7A501BDEQWKGSJT","yahoo-finance-forex","Fetches real-time FOREX news and market data from Yahoo Finance for major currency pairs (EUR\u002FUSD, GBP\u002FUSD, USD\u002FJPY, etc.). Analyzes sentiment and provides trading context.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fyahoo-finance-forex\u002FSKILL.md",[165],{"categoryNo":166,"categoryName":167,"taxonomyType":25,"taxonomyTypeDesc":26},"01JBSEEDCAT000000000000040","财务",[169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,34],"finance","yahoo finance","forex","yahoo","yahoo finance forex","github","https","sentiment","real-time","python3","install","Market","usd","news","time","currency","gbp","pip","market data",{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":192,"categories":193,"keywords":195,"publishedAt":48},"01M6Q5QM6VS84WZVM7BKXXPF32","yahoo-finance","This skill should be used when the user asks to \"get stock prices\", \"check stock quotes\", \"look up earnings\", \"get financial data\", \"find trending stocks\", or needs stock market data from Yahoo Finance.","https:\u002F\u002Fgithub.com\u002FKernel8901\u002Fai-agent-skills-classification\u002Fblob\u002Fmain\u002F金融\u002Fyahoo-finance-cli\u002FSKILL.md",[194],{"categoryNo":166,"categoryName":167,"taxonomyType":25,"taxonomyTypeDesc":26},[169,170,172,196,197,198,199,200,201,180,202,203,204,205,206,207,208,209,137,187],"cli","homebrew","find","check","brew","up","local","get","stock","link","trending","sf","quotes","ln"]