[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6FEHXW8Y4THGPM8P54F14DY":3,"skill-same:01M6FEHXW8Y4THGPM8P54F14DY":52},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":25,"linkOnly":26,"markdownContent":27,"markdownBody":27,"frontMatter":28,"refName":29,"commitSha":30,"skillDirectory":31,"forksCount":32,"repositoryHtmlUrl":33,"repositoryHomepageUrl":34,"ownerHtmlUrl":35,"ownerBlogUrl":6,"ownerDescription":36,"updatedAt":37,"install":38,"resources":51},"01M6FEHXW8Y4THGPM8P54F14DY","skill-authoring",null,"ComeOnOliver\u002Fskillshub","ComeOnOliver","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F43215937?v=4","01M6E5JNFECJAAVFE0W7SMWWD2","Haoran Shu","01M6E5JNNDQS7AJZDYFQ8TPZCK",67,"MIT","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fpproenca\u002Fdot-skills\u002Fskill-authoring\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[23,24],"skill authoring","authoring","2026-11-10 10:12:30",false,"# AI Agent Skills Best Practices\n\nDesign and development guide for AI agent skills, including Claude Code skills and MCP tools. Contains 46 rules across 8 categories, prioritized by impact to guide skill creation, review, and optimization.\n\n## When to Apply\n\n- Creating new Claude Code skills or MCP tools\n- Writing or reviewing SKILL.md metadata and descriptions\n- Optimizing skill trigger reliability\n- Structuring content for progressive disclosure\n- Testing skill activation and behavior\n- Designing tool interfaces for agent workflows\n\n## Core Principles\n\n**1. Descriptions drive activation.** Claude selects skills based on description matching against user intent. Include specific capabilities, trigger keywords, and negative cases. A skill with a vague description activates inconsistently or never.\n\n**2. Front-load critical instructions.** Claude may truncate long content. Place non-negotiable rules in the first 100 lines. Bury important constraints at the end and they get ignored.\n\n**3. Progressive disclosure saves tokens.** Load detailed content only when needed. A 2000-line skill wastes context on every activation. Structure as: SKILL.md (overview) → references\u002F (details) → scripts\u002F (execution).\n\n**4. Test activation, not just execution.** A skill that works perfectly but never triggers provides zero value. Test with real user phrases, synonyms, and edge cases before deployment.\n\n**5. One skill per domain.** Overlapping skills create activation conflicts. Split by clear boundaries (language, framework, workflow stage) with distinct trigger keywords.\n\n## Rule Categories\n\n| Priority | Category | Impact | Prefix |\n|----------|----------|--------|--------|\n| 1 | Skill Metadata Design | CRITICAL | `meta-` |\n| 2 | Description Engineering | CRITICAL | `desc-` |\n| 3 | Content Structure | HIGH | `struct-` |\n| 4 | Trigger Optimization | HIGH | `trigger-` |\n| 5 | Progressive Disclosure | MEDIUM-HIGH | `prog-` |\n| 6 | MCP Tool Design | MEDIUM | `mcp-` |\n| 7 | Testing and Validation | MEDIUM | `test-` |\n| 8 | Maintenance and Distribution | LOW-MEDIUM | `maint-` |\n\n## Quick Reference\n\n### 1. Skill Metadata Design (CRITICAL)\n\n- `meta-name-format` - Use lowercase hyphenated skill names\n- `meta-name-hyphen-boundaries` - Never start or end names with hyphens\n- `meta-name-no-consecutive-hyphens` - Avoid consecutive hyphens in names\n- `meta-name-uniqueness` - Ensure skill names are globally unique\n- `meta-required-frontmatter` - Include all required frontmatter fields\n- `meta-allowed-frontmatter-fields` - Use only allowed frontmatter fields\n- `meta-frontmatter-yaml-syntax` - Use valid YAML frontmatter syntax\n- `meta-name-length` - Keep skill names under 64 characters\n- `meta-directory-match` - Match skill name to directory name\n\n### 2. Description Engineering (CRITICAL)\n\n- `desc-specific-capabilities` - Name specific capabilities in description\n- `desc-trigger-keywords` - Include user trigger keywords in description\n- `desc-third-person-voice` - Write descriptions in third person\n- `desc-length-optimization` - Optimize description length for discovery\n- `desc-avoid-vague-terms` - Avoid vague terms in descriptions\n- `desc-differentiate-similar-skills` - Differentiate similar skills with distinct triggers\n- `desc-include-negative-cases` - Include negative cases for precision\n\n### 3. Content Structure (HIGH)\n\n- `struct-header-hierarchy` - Use consistent header hierarchy\n- `struct-instructions-first` - Put critical instructions early in content\n- `struct-imperative-instructions` - Write instructions in imperative mood\n- `struct-code-blocks-with-language` - Specify language in code blocks\n- `struct-line-limit` - Keep SKILL.md under 500 lines\n- `struct-single-responsibility` - One skill per domain\n\n### 4. Trigger Optimization (HIGH)\n\n- `trigger-slash-command-aliases` - Include slash command aliases in description\n- `trigger-file-type-patterns` - Include file type patterns in description\n- `trigger-workflow-stages` - Reference workflow stages in description\n- `trigger-error-patterns` - Include error patterns in debugging skills\n- `trigger-synonym-coverage` - Cover synonyms and alternate phrasings\n\n### 5. Progressive Disclosure (MEDIUM-HIGH)\n\n- `prog-three-level-disclosure` - Implement three-level progressive disclosure\n- `prog-one-level-deep-links` - Limit reference links to one level deep\n- `prog-scripts-execute-not-read` - Execute scripts instead of reading code\n- `prog-lazy-load-examples` - Lazy load examples and reference material\n- `prog-mutual-exclusion` - Separate mutually exclusive contexts\n\n### 6. MCP Tool Design (MEDIUM)\n\n- `mcp-tool-naming` - Use clear action-object tool names\n- `mcp-parameter-descriptions` - Document all tool parameters\n- `mcp-error-messages` - Return actionable error messages\n- `mcp-tool-scope` - Design single-purpose tools\n- `mcp-allowed-tools` - Use allowed-tools for safety constraints\n- `mcp-idempotent-operations` - Design idempotent tool operations\n\n### 7. Testing and Validation (MEDIUM)\n\n- `test-trigger-phrases` - Test skill activation with real user phrases\n- `test-edge-cases` - Test skills with edge case inputs\n- `test-negative-scenarios` - Test that skills do NOT trigger on unrelated requests\n- `test-instruction-clarity` - Test instructions with fresh context\n\n### 8. Maintenance and Distribution (LOW-MEDIUM)\n\n- `maint-semantic-versioning` - Use semantic versioning for skill releases\n- `maint-changelog` - Maintain a changelog for skill updates\n- `maint-plugin-packaging` - Package skills as plugins for distribution\n- `maint-audit-security` - Audit skills before installing from external sources\n\n## Creating Rules\n\nCopy [assets\u002Ftemplates\u002F_template.md](assets\u002Ftemplates\u002F_template.md) and follow the frontmatter schema:\n\n```yaml\n---\ntitle: Rule Title Here\nimpact: CRITICAL|HIGH|MEDIUM-HIGH|MEDIUM|LOW-MEDIUM|LOW\nimpactDescription: Quantified impact (e.g., \"2-10x improvement\")\ntags: prefix, technique, related-concepts\n---\n```\n\nReference files use the pattern: `references\u002F{prefix}-{slug}.md`\n\n## References\n\n- [skills-ref specification](https:\u002F\u002Fgithub.com\u002Fagentskills\u002Fagentskills\u002Ftree\u002Fmain\u002Fskills-ref)\n- [Anthropic Engineering: Agent Skills](https:\u002F\u002Fwww.anthropic.com\u002Fengineering\u002Fequipping-agents-for-the-real-world-with-agent-skills)\n- [Claude Code Skills Documentation](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fskills)\n- [Anthropic Skills Repository](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fskills)\n- [MCP Best Practices](https:\u002F\u002Fmodelcontextprotocol.info\u002Fdocs\u002Fbest-practices\u002F)\n- [Prompt Engineering Guide: LLM Agents](https:\u002F\u002Fwww.promptingguide.ai\u002Fresearch\u002Fllm-agents)\n- [Claude Skills Deep Dive](https:\u002F\u002Fleehanchung.github.io\u002Fblogs\u002F2025\u002F10\u002F26\u002Fclaude-skills-deep-dive\u002F)\n",{},"main","def8531e65114c0fca8fb8551c1871ee0eed705c","skills\u002Fpproenca\u002Fdot-skills\u002Fskill-authoring",24,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub","https:\u002F\u002Fskillshub.wtf","https:\u002F\u002Fgithub.com\u002FComeOnOliver","AI & Blockchain Engineer\r\n","2026-06-24 22:07:19",{"source":7,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add ComeOnOliver\u002Fskillshub --skill skill-authoring",{"tool":44,"command":45},"bunx","bunx skills add ComeOnOliver\u002Fskillshub --skill skill-authoring",{"tool":47,"command":48},"pnpm","pnpm dlx skills add ComeOnOliver\u002Fskillshub --skill skill-authoring","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fpproenca\u002Fdot-skills\u002Fskill-authoring\u002FSKILL.md",true,[],[53,63,73,82,88,95],{"skillNo":54,"name":55,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":59,"publishedAt":25},"01M6FKJN13GMG4KY57J3QD8GY5","model-hierarchy-skill","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzscole\u002Fmodel-hierarchy-skill\u002Fmodel-hierarchy-skill\u002FSKILL.md",[58],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[60,61,62],"model","model hierarchy","model hierarchy skill",{"skillNo":64,"name":65,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":69,"publishedAt":25},"01M6FKJJJ6GCAMR4PPPRD12M5X","zoom-mcp","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-mcp\u002FSKILL.md",[68],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[70,71,72],"MCP","zoom","zoom mcp",{"skillNo":74,"name":75,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":25},"01M6FKJE3KYT3KPEZMEEHYKA6K","zoom-apps-sdk","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[79,80,81,71],"sdk","zoom apps sdk","apps",{"skillNo":83,"name":84,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":25},"01M6FKJ9EDH9Q5VWDF8HAJRSP0","windows","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwindows\u002FSKILL.md",[],[84],{"skillNo":89,"name":90,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":94,"publishedAt":25},"01M6FKJ666BEEFC3AE9GJ72Q93","whiteboard","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwhiteboard\u002FSKILL.md",[93],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},[90],{"skillNo":96,"name":97,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":25},"01M6FKJ21WFCDGTJQ2Q91T9DVW","websockets","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwebsockets\u002FSKILL.md",[],[97]]