What exactly is broken in popular skills: references to files that are not bundled, actions taken without human approval, processes with no stopping condition. Everything here is measured from the skill text rather than judged by a model, so every line can be checked.
Generatorwill not run
End-to-end Xiaohongshu operations including positioning, topic research, content production, publish execution, and post-incident recovery. Reusable across verticals with templates
- References files that are not bundled: examples/<vertical>/<vertical>.md, examples/<vertical>/README.md, examples/lifestyle/
- 15 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (3 tags): a typed call is more reliable
clawhub · ↓ 733
GeneratorObsidianwill not run
Capture social media links (Weibo, Xiaohongshu, WeChat, Xiaoyuzhou) — extract text, images, and metadata, then generate a Markdown note with AI deep summary, saved to the user's Ob
- References files that are not bundled: attachments/{date}-{slug}/img-001.jpg, original_url, attachments/{date}-{slug}/img-NNN.jpg
- 15 top-level sections: this looks like several domains in one skill
- 6 referenced file(s) missing: attachments/{date}-{slug}/img-001.jpg, original_url, attachments/{date}-{slug}/img-NNN.jpg
clawhub · ↓ 734
Integrationwill not run
Generate and edit images with Qwen Image via DashScope API. This is a skill, not a callable tool. First use the read tool to open this SKILL.md, then run the script it specifies; n
- References files that are not bundled: ...
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: ...
clawhub · ↓ 733
Analyzerwill not run
Operate, verify, rebuild, and debug the actual MemoryLab long-term memory sidecar that feeds `memory/ACTIVE_TASK_STATE.md` and `memory/LIVE_CONTEXT_PACKET.md`. Use when working on
- References files that are not bundled: scripts/context-refresh*
- 1 referenced file(s) missing: scripts/context-refresh*
- Does not say what the result is
clawhub · ↓ 734
GeneratorShopifywill not run
Bulk upload products to Shopify stores. Read product data from Excel/CSV, automatically create products, images, variants, prices and inventory. Use cases: (1) Batch list new produ
- References files that are not bundled: assets/products.xlsx, assets/products.csv, scripts/config.py
- 3 referenced file(s) missing: assets/products.xlsx, assets/products.csv, scripts/config.py
- No condition that starts the skill
clawhub · ↓ 759
Integrationwill not run
Unified multi-engine web search. Use when the user wants to search the web, find information, look up sources, or perform research. Supports Tavily API (fast) and browser-based sea
- References files that are not bundled: url, ../agent-browser/SKILL.md
- 2 referenced file(s) missing: url, ../agent-browser/SKILL.md
- Does not say what the process needs to start
clawhub · ↓ 710
Procedurewill not run
An autonomous trading agent for Polymarket (Polygon). Scans 15-minute markets for momentum and trades automatically. Includes dashboard, simulation mode, and live execution.
- References files that are not bundled: scripts/bootstrap.sh, scripts/config.yaml.example
- 2 referenced file(s) missing: scripts/bootstrap.sh, scripts/config.yaml.example
- Does not say what the result is
clawhub · ↓ 733
Integrationwill not run
A Google Play Store API alternative on fetcher.sh — pay-per-call in USDC via x402, or prepaid credits with a Bearer key, no Google Play Console access. Use when the user wants to s
- References files that are not bundled: ../fetcher/SKILL.md
- 1 referenced file(s) missing: ../fetcher/SKILL.md
- Does not say what the result is
clawhub · ↓ 2 738
Procedurewill not run
Generate songs, instrumentals, or lyrics-driven tracks through a structured OpenClaw-native workflow with anti-sparse prompt engineering and quality verification. Provider-agnostic
- References files that are not bundled: /Users/luis/Repos/skills/publish/LOCAL-MODELS-INSTALL-GUIDE.md, ../music-craft-minimax/scripts/lint_music_request.py, ../music-craft-minimax/references/emotion-delivery.md
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 26 top-level sections: this looks like several domains in one skill
clawhub · ↓ 983
Procedurewill not run
团队预算管理工具 - 帮助小型团队规划和追踪项目预算
- References files that are not bundled: examples/sample-budget.md, templates/budget-template.json
- 2 referenced file(s) missing: examples/sample-budget.md, templates/budget-template.json
- Does not say what the result is
clawhub · ↓ 757
GeneratorNotionwill not run
This skill should be used when writing blog posts, articles, or long-form content in the writer's distinctive writing style. It produces authentic, opinionated content that matches
- References files that are not bundled: references/blog-examples/, references/style-guide.md
- 2 referenced file(s) missing: references/blog-examples/, references/style-guide.md
- No condition that starts the skill
clawhub · ↓ 785
ProcedureGitHubwill not run
把已有的 Word、PDF 或 PPT,直接制作成可在浏览器中汇报、带分步动效、支持全屏展示与离线交付的 16:9 HTML 演示文稿。它不只是转换文件格式,还会重新设计页面、视觉层级和讲解节奏,让成品可以直接用于项目汇报、客户提案、产品路演或培训。如果你准备写报告、做汇报或制作提案,但目前只有一个主题、还没有完整思路,TaoHtml 也可以通过少量关键问题
- References files that are not bundled: scripts/preflight.py, references/agent-workflow.md, references/...
- 5 referenced file(s) missing: scripts/preflight.py, references/agent-workflow.md, references/...
- Does not say what the result is
clawhub · ↓ 967 ★ 1
Referencewill not run
- References files that are not bundled: ../../workspace/AGENTS.md, ../../workspace/MEMORY.md, ../../workspace/IDENTITY.md
- 10 top-level sections: this looks like several domains in one skill
- 14 referenced file(s) missing: ../../workspace/AGENTS.md, ../../workspace/MEMORY.md, ../../workspace/IDENTITY.md
clawhub · ↓ 732
Personawill not run
基于DMP标签体系的用户画像洞察分析技能。支持多维度数据分析、核心特征筛选、推导说明生成、策略建议输出,适用于精准营销、用户运营、产品优化等场景。当用户需要分析用户画像数据、生成用户洞察报告、制定营销策略时使用此技能。触发词:用户画像、人群画像、画像洞察、画像分析、画像报告、人群洞察、洞察分析、洞察报告、TGI分析、标签分析、特征分析、人群特征、营销策略、精
- References files that are not bundled: references/ppt-generation-guide.md
- 1 referenced file(s) missing: references/ppt-generation-guide.md
- Does not say what the result is
clawhub · ↓ 731
Analyzerwill not run
This skill should be used when the user asks to "implement LLM-as-judge", "compare model outputs", "create evaluation rubrics", "mitigate evaluation bias", or mentions direct scori
- References files that are not bundled: references/implementation-patterns.md, references/bias-mitigation.md, references/metrics-guide.md
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 3 referenced file(s) missing: references/implementation-patterns.md, references/bias-mitigation.md, references/metrics-guide.md
clawhub · ↓ 731
Procedurewill not run
OpenClaw 智能度综合评伌技能。围绕 14 个维度(含规划能力、幻觉控制)输出综合评分、证据、风险与趋势。对齐 CLEAR/T-Eval/Anthropic 行业标准。
- References files that are not bundled: scripts/regression-metrics-report.py
- The response is described with custom markup (4 tags): a typed call is more reliable
- 1 referenced file(s) missing: scripts/regression-metrics-report.py
clawhub · ↓ 731
Generatorwill not run
Create premium frontend interfaces with modern design systems. Forces generation of design.md for consistency, follows 2026 trends, ensures accessibility, and provides quality chec
- References files that are not bundled: references/a11y-checklist.md, references/trends-2026.md, references/tech-stack.md
- 12 top-level sections: this looks like several domains in one skill
- 3 referenced file(s) missing: references/a11y-checklist.md, references/trends-2026.md, references/tech-stack.md
clawhub · ↓ 756
Procedurewill not run
获取B站、抖音等平台当天的热门梗/迷因,追溯热梗来源、解释含义、整理相关视频素材,生成结构化的热梗百科文档。适用于网络文化研究、内容创作灵感、社交话题追踪等场景。
- References files that are not bundled: {链接}, {URL}
- 2 referenced file(s) missing: {链接}, {URL}
- Does not say what the result is
clawhub · ↓ 730
Procedurewill not run
Decompose complex physical tasks into atomic subtasks for robot execution.
Use when user provides: (1) An image showing a physical scene (indoor/outdoor), and (2) A natural languag
- References files that are not bundled: action-bank.md, examples.md
- 2 referenced file(s) missing: action-bank.md, examples.md
- Linear process with no failure handling
clawhub · ↓ 730
Integrationwill not run
跨平台统计软件 CLI 集成,面向 AI Agent;覆盖 34+ 款软件(R/Stata/SAS/SPSS/Python/贝叶斯/ML等),双语。核心价值:激活历史代码资产,用于 AI 工作流自动化。 / Cross-platform statistical software CLI integration for AI Agent; 34+ packag
- References files that are not bundled: references/*.md
- 10 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: references/*.md
clawhub · ↓ 2 723
ProcedurePlaywrightwill not run
Auto-publish Zhihu column articles on Linux (Chrome attach/VNC). Agent writes title+body then exec zhihu_publish.sh with --publish --submit. Triggers 知乎发帖, 发知乎, 知乎文章, 写一篇发知乎, 帮我发帖,
- References files that are not bundled: scripts/chromedriver.env, scripts/screenshots/
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 10 top-level sections: this looks like several domains in one skill
clawhub · ↓ 782
Integrationwill not run
易经占卜大师——调用李小问 API 起卦、解卦,支持铜钱(默认)/大衍起卦,也支持用户自报卦象。两种渲染风格:图文(透传 Markdown)/ 艺术字(终端 ASCII 卦图+评级框)。用户涉及占卜、算卦、起卦、摇卦、解卦、易经、周易、六爻、卦象、问卦、求签、测一卦等意图时触发。
- References files that are not bundled: url
- The response is described with custom markup (5 tags): a typed call is more reliable
- 1 referenced file(s) missing: url
clawhub · ↓ 729
Generatorwill not run
Create, optimize, or update AgentSkills with evidence-based design. Two modes: (1) CREATE — build new skills from scratch with proven design patterns. (2) OPTIMIZE — diagnose and f
- References files that are not bundled: references/schemas.md, references/troubleshooting.md, references/api.md
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 11 referenced file(s) missing: references/schemas.md, references/troubleshooting.md, references/api.md
clawhub · ↓ 728
Procedurewill not run
- References files that are not bundled: references/composition-guide.md, references/lighting-guide.md, references/color-theory.md
- 4 referenced file(s) missing: references/composition-guide.md, references/lighting-guide.md, references/color-theory.md
- Does not say what the result is
clawhub · ↓ 729
Analyzerwill not run
- References files that are not bundled: templates/review_prompt.txt
- 12 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: templates/review_prompt.txt
clawhub · ↓ 728