[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-50cbd94d-75c5-440c-9aaf-9d457e2fe2bb":3,"$fAYTwHDNbDIpmsHJuu-Xk-cXpbz8Ot1V2e10Hr8r8RUQ":42},{"id":4,"title":5,"description":6,"categoryId":7,"moduleId":8,"tags":9,"prompt":10,"icon":11,"source":12,"sourceUrl":13,"authorId":14,"authorName":15,"isPublic":16,"stars":17,"runs":18,"createdAt":19,"updatedAt":19,"module":20,"category":27,"packages":33},"50cbd94d-75c5-440c-9aaf-9d457e2fe2bb","skill-improver","迭代改进Claude代码技能，使用技能审查代理直到达到质量标准。用于改进存在多个质量问题的技能、迭代新技能直到达到标准，或使用自动修复审查周期代替手动编辑。","cat_coding_backend","mod_coding","sickn33,coding","---\nname: skill-improver\ndescription: \"Iteratively improve a Claude Code skill using the skill-reviewer agent until it meets quality standards. Use when improving a skill with multiple quality issues, iterating on a new skill until it meets standards, or automated fix-review cycles instead of manual editing.\"\nrisk: unknown\nsource: community\n---\n\n# Skill Improvement Methodology\n\nIteratively improve a Claude Code skill using the skill-reviewer agent until it meets quality standards.\n\n## Prerequisites\n\nRequires the `plugin-dev` plugin which provides the `skill-reviewer` agent.\n\nVerify it's enabled: run `\u002Fplugins` — `plugin-dev` should appear in the list. If missing, install from the Trail of Bits plugin repository.\n\n## Core Loop\n\n1. **Review** - Call skill-reviewer on the target skill\n2. **Categorize** - Parse issues by severity\n3. **Fix** - Address critical and major issues\n4. **Evaluate** - Check minor issues for validity before fixing\n5. **Repeat** - Continue until quality bar is met\n\n## When to Use\n- Improving a skill with multiple quality issues\n- Iterating on a new skill until it meets standards\n- Automated fix-review cycles instead of manual editing\n- Consistent quality enforcement across skills\n\n## When NOT to Use\n\n- **One-time review**: Use `\u002Fskill-reviewer` directly instead\n- **Quick single fixes**: Edit the file directly\n- **Non-skill files**: Only works on SKILL.md files\n- **Experimental skills**: Manual iteration gives more control during exploration\n\n## Issue Categorization\n\n### Critical Issues (MUST fix immediately)\n\nThese block skill loading or cause runtime failures:\n\n- Missing required frontmatter fields (name, description) — Claude cannot index or trigger the skill\n- Invalid YAML frontmatter syntax — Parsing fails, skill won't load\n- Referenced files that don't exist — Runtime errors when Claude follows links\n- Broken file paths — Same as above, leads to tool failures\n\n### Major Issues (MUST fix)\n\nThese significantly degrade skill effectiveness:\n\n- Weak or vague trigger descriptions — Claude may not recognize when to use the skill\n- Wrong writing voice (second person \"you\" instead of imperative) — Inconsistent with Claude's execution model\n- SKILL.md exceeds 500 lines without using references\u002F — Overloads context, reduces comprehension\n- Missing \"When to Use\" or \"When NOT to Use\" sections — Required by project quality standards\n- Description doesn't specify when to trigger — Skill may never be selected\n\n### Minor Issues (Evaluate before fixing)\n\nThese are polish items that may or may not improve the skill:\n\n- Subjective style preferences — Reviewer may have different taste than author\n- Optional enhancements — May add complexity without proportional value\n- \"Nice to have\" improvements — Consider cost-benefit before implementing\n- Formatting suggestions — Often valid but low impact\n\n## Minor Issue Evaluation\n\nBefore implementing any minor issue fix, evaluate:\n\n1. **Is this a genuine improvement?** - Does it add real value or just satisfy a preference?\n2. **Could this be a false positive?** - Is the reviewer misunderstanding context?\n3. **Would this actually help Claude use the skill?** - Focus on functional improvements\n\nOnly implement minor fixes that are clearly beneficial. Skill-reviewer may produce false positives.\n\n## Invoking skill-reviewer\n\nUse the skill-reviewer agent from the plugin-dev plugin. Request a review by asking Claude to:\n\n> Review the skill at [SKILL_PATH] using the plugin-dev:skill-reviewer agent. Provide a detailed quality assessment with issues categorized by severity.\n\nReplace `[SKILL_PATH]` with the absolute path to the skill directory (e.g., `\u002Fpath\u002Fto\u002Fplugins\u002Fmy-plugin\u002Fskills\u002Fmy-skill`).\n\n## Example Fix Cycle\n\n**Iteration 1 — skill-reviewer output:**\n```text\nCritical: SKILL.md:1 - Missing required 'name' field in frontmatter\nMajor: SKILL.md:3 - Description uses second person (\"you should use\")\nMajor: Missing \"When NOT to Use\" section\nMinor: Line 45 is verbose\n```\n\n**Fixes applied:**\n- Added name field to frontmatter\n- Rewrote description in third person\n- Added \"When NOT to Use\" section\n\n**Iteration 2 — run skill-reviewer again to verify fixes:**\n```text\nMinor: Line 45 is verbose\n```\n\n**Minor issue evaluation:**\nLine 45 communicates effectively as-is. The verbosity provides useful context. Skip.\n\n**All critical\u002Fmajor issues resolved. Output the completion marker:**\n```\n\u003Cskill-improvement-complete>\n```\n\nNote: The marker MUST appear in the output. Statements like \"quality bar met\" or \"looks good\" will NOT stop the loop.\n\n## Completion Criteria\n\n**CRITICAL**: The stop hook ONLY checks for the explicit marker below. No other signal will terminate the loop.\n\nOutput this marker when done:\n\n```\n\u003Cskill-improvement-complete>\n```\n\n**When to output the marker:**\n\n1. **skill-reviewer reports \"Pass\"** or **no issues found** → output marker immediately\n2. **All critical and major issues are fixed** AND you've verified the fixes → output marker\n3. **Remaining issues are only minor** AND you've evaluated them as false positives or not worth fixing → output marker\n\n**When NOT to output the marker:**\n\n- Any critical issue remains unfixed\n- Any major issue remains unfixed\n- You haven't run skill-reviewer to verify your fixes worked\n\nThe marker is the ONLY way to complete the loop. Natural language like \"looks good\" or \"quality bar met\" will NOT stop the loop.\n\n## Rationalizations to Reject\n\n- \"I'll just mark it complete and come back later\" - Fix issues now\n- \"This minor issue seems wrong, I'll skip all of them\" - Evaluate each one individually\n- \"The reviewer is being too strict\" - The quality bar exists for a reason\n- \"It's good enough\" - If there are major issues, it's not good enough\n\n## Limitations\n- Use this skill only when the task clearly matches the scope described above.\n- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.\n- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.\n","","imported","https:\u002F\u002Fgithub.com\u002Fsickn33\u002Fantigravity-awesome-skills","user_system_seed","SkillOPIC",true,234,1353,"2026-05-16 13:40:47",{"id":8,"name":21,"slug":22,"icon":23,"description":24,"sort":25,"createdAt":26},"编程开发","coding","mdi-code-braces","代码生成、调试、审查，提升开发效率",2,"2026-05-16 12:53:40",{"id":7,"name":28,"slug":29,"icon":30,"description":31,"moduleId":8,"sort":25,"skillCount":32,"createdAt":26},"后端开发","backend","mdi-server","API、数据库、服务端架构",296,[34],{"id":35,"skillId":4,"version":36,"fileName":37,"fileSize":38,"filePath":39,"fileHash":40,"manifest":41,"createdAt":19},"19ee6493-be77-43e6-9301-7e21fe13368c","1.0.0","skill-improver.zip",2680,"uploads\u002Fskills\u002F50cbd94d-75c5-440c-9aaf-9d457e2fe2bb\u002Fskill-improver.zip","a156e7c7dc88b9871daf106111e7d41a6b4f9347bdda6bdf507e459219ac9355","[{\"path\":\"SKILL.md\",\"isDirectory\":false,\"size\":6074}]",{"code":43,"message":44,"data":45},200,"success",{"items":46,"stats":47,"page":50},[],{"averageRating":48,"totalRatings":48,"ratingCounts":49},0,[48,48,48,48,48],{"limit":51,"offset":48,"hasMore":52,"nextOffset":51,"ratedOnly":16},15,false]