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
PRD文档生成器(团队版) - 从需求描述一键生成完整PRD文档与HTML交互原型,支持iWiki发布
- References files that are not bundled: images/xx.png, images/xxx.png, images/{页面A}.png
- 11 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (24 tags): a typed call is more reliable
clawhub · ↓ 858 ★ 2
Analyzerwill not run
Supply chain intelligence for OpenClaw skills. Use when analyzing the local skill ecosystem, understanding tool and package dependencies, discovering skill categories, mapping rela
- References files that are not bundled: scripts/*.py, scripts/ontology.py
- 2 referenced file(s) missing: scripts/*.py, scripts/ontology.py
- Does not say what the result is
clawhub · ↓ 835 ★ 2
Procedurewill not run
通过云端连接器优先执行浏览器插件数据采集,也可回退到本地连接器;适用于抖音、TikTok、小红书、Amazon、Bilibili 的采集任务,以及 web-collection 首次上手、配置、付费使用说明和 QA 排障。
- References files that are not bundled: assets/bitable-step-00-personal-center.png, assets/bitable-step-01-template-copy.png, assets/bitable-step-02-auth-code.png
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 17 top-level sections: this looks like several domains in one skill
clawhub · ↓ 2 329
Templatewill not run
Use OpenClaw as a supervisor-driven orchestration scaffold for local coding CLIs such as Codex, Claude Code, and OpenCode. Supports task records, lifecycle transitions, worker laun
- References files that are not bundled: assets/scripts/run-codex.ps1, assets/scripts/run-claude-code.ps1, assets/scripts/run-opencode.ps1
- 24 top-level sections: this looks like several domains in one skill
- 12 referenced file(s) missing: assets/scripts/run-codex.ps1, assets/scripts/run-claude-code.ps1, assets/scripts/run-opencode.ps1
clawhub · ↓ 1 203 ★ 1
Analyzerwill not run
Search and discover trending products from ProductHunt with structured analysis data. Use when users ask about product recommendations, tool comparisons, or trending apps.
- References files that are not bundled: phUrl, websiteUrl
- 2 referenced file(s) missing: phUrl, websiteUrl
- Does not say what the process needs to start
clawhub · ↓ 2 164
Procedure
Navigate Uber Eats in a live browser or app handoff to compare merchants, manage carts, and reach checkout safely.
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 14 top-level sections: this looks like several domains in one skill
- The skill tells the model to perform an irreversible action with no human approval
clawhub · ↓ 832 ★ 2
Procedurewill not run
腾讯混元生图API - 根据文本描述生成AI图像
- References files that are not bundled: README.md
- 1 referenced file(s) missing: README.md
- Does not say what the result is
clawhub · ↓ 1 177 ★ 1
Procedurewill not run
- References files that are not bundled: [a-zA-Z0-9-]{0,61}[a-zA-Z0-9]
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: [a-zA-Z0-9-]{0,61}[a-zA-Z0-9]
clawhub · ↓ 2 182
Integration
Manage tasks and projects using the TaskBoardAI Kanban system. Includes MCP server integration.
- The response is described with custom markup (10 tags): a typed call is more reliable
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 141
GeneratorWord
Use this skill any time a .docx file is involved -- as input, output, or both. This includes: creating Word documents, reports, letters, memos, or proposals; reading, parsing, or e
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 12 top-level sections: this looks like several domains in one skill
- The skill tells the model to perform an irreversible action with no human approval
clawhub · ↓ 1 293 ★ 1
Procedure
Professional KuCoin (KC) trading system - multi-account support, spot/margin/futures trading, asset transfers. Use to check balances, transfer assets, open/close positions, and man
- 10 top-level sections: this looks like several domains in one skill
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 176 ★ 1
Analyzerwill not run
botlearn-healthcheck — BotLearn autonomous health inspector for OpenClaw instances across 5 domains (hardware, config, security, skills, autonomy); triggers on system check, health
- References files that are not bundled: scripts/collect-*.sh
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 11 top-level sections: this looks like several domains in one skill
clawhub · ↓ 1 198 ★ 1
Integrationwill not run
Direct iCloud Calendar integration via CalDAV protocol. Create, read, update, and delete calendar events without third-party services. Use when the user wants to manage their iClou
- References files that are not bundled: references/icloud-endpoints.md
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: references/icloud-endpoints.md
clawhub · ↓ 2 215
Generatorwill not run
Patterns for creating design systems with personality and distinctive aesthetics. Covers aesthetic documentation, color token architecture, typography systems, layered surfaces, an
- References files that are not bundled: ../../meta/design-system-creation/, ../design-system-components/, ../loading-state-patterns/
- 13 top-level sections: this looks like several domains in one skill
- 3 referenced file(s) missing: ../../meta/design-system-creation/, ../design-system-components/, ../loading-state-patterns/
clawhub · ↓ 2 193
Referencewill not run
Switch between AI models dynamically to optimize costs and performance. Use when the user says mode commands like "eco mode", "balanced mode", "smart mode", or "max mode", or when
- References files that are not bundled: references/REFERENCE.md
- 1 referenced file(s) missing: references/REFERENCE.md
- Does not say what the result is
clawhub · ↓ 2 192
Integration
多机器人协同控制 Skill。让 AI Agent 具备感知反馈、动态适配、并行调度多种机器人的能力。支持机械臂、四足机器人等任意 HTTP API 机器人。
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 855 ★ 2
ProcedurePlaywright
Browser-automated TikTok streak messaging skill using Playwright. Sends daily messages to configured usernames with state tracking, retry control, and optional content discovery vi
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 664 ★ 1
Procedurewill not run
火车票智能中转方案推荐助手。查火车票/高铁票/动车票/12306余票,智能拼接中转换乘方案,学习用户偏好(坐席策略、时间偏好、中转约束)。火车票查询/余票/车次/预订/中转换乘/坐席偏好。
- References files that are not bundled: {jumpUrl}
- 1 referenced file(s) missing: {jumpUrl}
- Does not say what the result is
clawhub · ↓ 853 ★ 2
Integrationwill not run
百度地图 iOS SDK 与 BMKLocationKit 集成与开发规范。覆盖地图、定位、步骑行导航、检索、路线、标注与覆盖物;输出专业地图方案。可快捷使用百度地图 SDK 的能力与数据,构建功能丰富、交互性强的专业地图类应用。
- References files that are not bundled: assets/images/
- 1 referenced file(s) missing: assets/images/
- Does not say what the result is
clawhub · ↓ 1 263 ★ 1
Generatorwill not run
Embeds and controls a virtual avatar using the avatar-runtime npm package. Provides Live2D rendering, VRM 3D, vector fallback, and control-driven expression/body/scene animation vi
- References files that are not bundled: scripts/ensure-default-vrm-sample.sh, scripts/ensure-default-live2d-sample.sh, assets/vrm/slot/default.vrm
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 15 top-level sections: this looks like several domains in one skill
clawhub · ↓ 1 194 ★ 1
Reference
使用 ScholarGraph v1.4 及以上版本 进行癌症融合基因文献调研的全流程技能。包括搜索、多轮合并去重、提取信息、生成报告和 Excel 表格。
- 10 top-level sections: this looks like several domains in one skill
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 828 ★ 2
Analyzerwill not run
botlearn-assessment — BotLearn 5-dimension capability self-assessment (reasoning, retrieval, creation, execution, orchestration); triggers on botlearn assessment, capability test,
- References files that are not bundled: references/d{N}-q{L}-{difficulty}.md
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: references/d{N}-q{L}-{difficulty}.md
clawhub · ↓ 1 172 ★ 1
GeneratorYouTube
Video to text converter. Downloads videos from Bilibili using bilibili-api, from other sites using yt-dlp, then transcribes audio using faster-whisper. Use when you need to convert
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 174 ★ 1
Procedurewill not run
把一本书里沉淀的方法论蒸馏成一组原子化、可被Agent调用的skills,同时产出给人看的笔记和话术库。触发:用户说'帮我拆《XX》''把XX蒸馏成skill''distill this book into skills'时。不做书摘/读后感/作者人设角色扮演。
- References files that are not bundled: templates/BOOK_OVERVIEW.md.template, templates/SKILL.md.template, templates/INDEX.md.template
- 10 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (4 tags): a typed call is more reliable
clawhub · ↓ 1 914 ★ 1
Procedure
Fetches skills.sh trending rankings. Use when asking about skill rankings or popular tools.
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 246