Skill Seo — 技能 Seo

审计 and improve the tar获取 技能 for discovery, 搜索 recall, and 命令行工具ck-through across 技能 directories.

Prioritize ClawHub and 技能s.sh. Treat OpenClaw, Claude Code, Codex, and Cursor as compatibility constrAInts, not primary optimization surfaces.

Quick 启动

Use this 工作流 when you need a fast 审计 before publication:

运行 python3 {baseDir}/scripts/analyze_技能_seo.py <技能-path>. Read the tar获取 技能's name, description, first screen, and example prompts. 检查 whether the 技能 is easy to understand on a 列出ing page in under 10 seconds. Rewrite the highest-impact metadata and first-screen copy first. Re-运行 the 分析器 and compare the 结果 agAInst the baseline. When to Use This 技能

Use this 技能 when the goal is to make a 技能 easier to discover, understand, and trust on ClawHub, 技能s.sh, or similar directories.

Typical use cases:

Rewrite name, slug, or description for better 搜索 recall Improve 技能.md first-screen copy for higher 命令行工具ck-through Expand keyword, synonym, and user-intent coverage before publication 审计 whether a 技能 is ready for ClawHub or 技能s.sh submission Compare a 列出ing agAInst directory ranking 签名als and identify weak spots When Not to Use This 技能

Do not use this 技能 when the problem is primarily implementation 质量 rather than 列出ing discoverability.

Use a different 工作流 when you need to:

调试 运行time 失败s inside the tar获取 技能 Test whether the 技能 actually works end-to-end 添加 new product capabilities unrelated to discovery or 列出ing 质量 Build 平台-specific adapters unless the user explicitly asks for them If the 审计 Is Inconclusive

If the static 审计 is not enough, say exactly what is still uncertAIn and what evidence is missing.

Common next steps:

Read the tar获取 技能's full references and scripts for hidden termino记录y Inspect the public 列出ing page to compare displayed copy agAInst the local 技能.md Ask for real 搜索 terms, tar获取 audience, or competitor 列出ings 运行 a follow-up 质量 pass with 技能-test if the issue may be implementation rather than SEO Example Prompts 审计 this 技能 for ClawHub 搜索 visibility and rewrite the description for better recall. Why is this 技能 hard to discover on 技能s.sh, and what copy should I change first? 优化 this new 技能's slug, name, and top-of-file wording before I publish it. Compare this 技能 agAInst ClawHub and 技能s.sh ranking 签名als and give me a prioritized fix 列出. 审计 检查列出

Review the tar获取 技能 agAInst this 检查列出 before making changes:

Is the slug literal, 搜索able, and stable? Does the frontmatter description 状态 the job, object, outcome, and Use when ... trigger? Does the first screen show rea列出ic example prompts early? Are exact task phrases, synonyms, and user-intent phrases all present? Are prerequisites, constrAInts, and trust 签名als visible? Does the 技能 look credible and mAIntAIned rather than generic or template-like? Does the copy still match the actual capability boundary of the 技能? Are any optional extras, such as eval.yaml or UI metadata, actually relevant to the user's goal before recommending them? 工作流 Identify the tar获取 技能 folder and inspect 技能.md first. 运行 python3 {baseDir}/scripts/analyze_技能_seo.py <技能-path> to 获取 a baseline 报告. Read references/平台-签名als.md and references/optimization-patterns.md before making recommendations. If present, inspect README.md, example files, and any public 列出ing metadata that affects display. Separate recommendations by 平台: ClawHub: prioritize semantic recall, exact slug or name matches, examples, and popularity 签名als. 技能s.sh: prioritize clear category fit, high-conversion 列出ing copy, trust 签名als, and 安装-friendly presentation. 检查 that the 技能 remAIns broadly usable in OpenClaw, Claude Code, Codex, and Cursor without 添加ing 平台-specific optimization unless the user asks for it. Rewrite only what improves discovery or trigger 质量. Preserve the 技能's actual capability boundaries. If the user asks for implementation, edit the tar获取 技能 and re-运行 the 分析器 to confirm improvement. Definition of Done The 分析器 has 运行 and a baseline 报告 exists. All high severity findings have been 添加ressed or explicitly justified. The description covers at least 3 查询 classes: exact task phrase, synonyms, and user-intent language. Example prompts in 技能.md reflect rea列出ic user phrasing. Tar获取 平台 compatibility 状态 is explicit: verified where evidence exists, otherwise unverified. Re-运行ning the 分析器 shows improvement over the baseline. 输出 Requirements

Produce a concise 报告 with:

Current strengths and weaknesses 平台-specific findings for ClawHub and 技能s.sh Compatibility notes for OpenClaw, Claude Code, Codex, and Cursor using verified / unverified language only when relevant Missing keywords, synonyms, and user-phrased queries Recommended rewrites for name, sl