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.
Procedurewill not run
- References files that are not bundled: LICENSE
- 12 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: LICENSE
clawhub · ↓ 2 295 ★ 1
Reference
Book computer-repair services through Lokuli MCP. Use when user needs to find and book computer-repair. Triggers on requests like "book a computer-repair", "find computer-repair ne
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 308 ★ 1
Integrationwill not run
CLI for AI agents to find breweries for their humans. Uses Open Brewery DB. No auth required.
- References files that are not bundled: scripts/breweries
- 1 referenced file(s) missing: scripts/breweries
- Does not say what the process needs to start
clawhub · ↓ 2 333 ★ 1
Procedure
- 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 · ↓ 513 ★ 13
Procedurewill not run
生成HTML演示文稿/幻灯片 — 22 种风格模版,播放/演讲者模式。适用于路演、产品发布、技术分享、方案宣讲等场景。
- References files that are not bundled: references/brief-template.json, references/*.md, references/title-quality.md
- The response is described with custom markup (6 tags): a typed call is more reliable
- 13 referenced file(s) missing: references/brief-template.json, references/*.md, references/title-quality.md
clawhub · ↓ 2 429 ★ 1
Procedure
- 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 · ↓ 528 ★ 12
Generator
Generate conventional commit messages from code changes or diff.
- The response is described with custom markup (7 tags): a typed call is more reliable
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 614 ★ 9
Integration
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 322 ★ 1
Procedurewill not run
Complete memory system combining Baidu Embedding auto-recall, Git-Notes structured memory, and file-based workspace search. Use when setting up comprehensive agent memory with loca
- References files that are not bundled: scripts/file-search.sh
- 14 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: scripts/file-search.sh
clawhub · ↓ 2 249 ★ 1
Referencewill not run
Use this skill when the user wants quick, no-credential browser automation against public websites (summaries, keyword checks, simple page facts). Runs the local OpenClaw Automatio
- References files that are not bundled: examples/public_page_check
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: examples/public_page_check
clawhub · ↓ 1 585 ★ 2
ProcedureGitHubwill not run
观远 BI(Guandata)实战增益层。标准查数、指标批量查询、本地分析、建卡发布、ETL、数据集和表单结构管理、工作流、指标写入路由官方六组件。专攻 ETL 整库治理、SmartETL 全链路重写、引擎报错、自定义图表与 HTML 看板排障、v7 发布兼容、移动端 phoneLayout、SuperApp 的 LLM 中转及历史兼容、AI-native
- References files that are not bundled: scripts/patch_selector_linkage.js
- 37 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (15 tags): a typed call is more reliable
clawhub · ↓ 3 227 ★ 1
IntegrationGitHub
Manage Google Tasks from the command line - view, create, update, delete tasks and task lists. Use when the user asks to interact with Google Tasks, manage to-do items, create task
- 11 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
- Says nothing while it works
clawhub · ↓ 2 272 ★ 1
ProcedurePlaywrightwill not run
获取干净、可读的网页正文内容,适合现代网页、博客、新闻、公告和微信公众号文章抓取;支持网页正文提取、内容清洗、去噪、Markdown 输出,适用于普通 fetch 效果不佳、页面噪音较多或动态渲染干扰的场景。Clean content fetch for modern web pages, article extraction, WeChat article
- References files that are not bundled: scripts/fetch-web-content
- 1 referenced file(s) missing: scripts/fetch-web-content
- Does not say what the result is
clawhub · ↓ 1 572 ★ 2
Procedure
- 12 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 · ↓ 1 572 ★ 2
Reference
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 267 ★ 1
Analyzer
Recognizes various poses such as standing, sitting, lying down, bending, raising hands, running, and falling. Supports abnormal pose recognition and fall warnings, suitable for sec
- 1 steps
- Does not say what the result is
clawhub · ↓ 1 717 ★ 3
Integrationwill not run
Extends the org-cli skill so the agent also persists its own memory (knowledge, observations, daily notes) to an org workspace.
- References files that are not bundled: ../org-cli/SKILL.md
- 1 referenced file(s) missing: ../org-cli/SKILL.md
- Does not say what the result is
clawhub · ↓ 2 317 ★ 1
Generator
Temporary email helper backed by Vortex (vortex.email). Use when needing disposable addresses for signup flows: create a mailbox (random localpart), poll for messages, fetch and cl
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 239 ★ 1
Procedure
- The response is described with custom markup (4 tags): a typed call is more reliable
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 253 ★ 1
Analyzer
Verify the bead daemon is alive and responsive
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 248 ★ 1
Reference
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 258 ★ 1
Analyzerwill not run
Manage PARA-based personal knowledge management (PKM) systems using Projects, Areas, Resources, and Archives organization method. Use when users need to (1) Create a new PARA knowl
- References files that are not bundled: assets/AGENTS.md.template, assets/project.md.template, assets/area-overview.md.template
- 11 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 · ↓ 2 237 ★ 1
Integrationwill not run
8004 Agent Skill for registering AI agents on the ERC-8004 Trustless Agents standard and authenticating them via SIWA (Sign In With Agent). Use this skill when an agent needs to: (
- References files that are not bundled: assets/MEMORY.md.template
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: assets/MEMORY.md.template
clawhub · ↓ 2 290 ★ 1
Procedure
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 253 ★ 1
GeneratorDockerCloudflare
Browser automation — setup the bsession environment, fetch info from a website (one-shot), create scripted automations (one-shot or recurring), or debug existing sessions. Works fr
- 12 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (12 tags): a typed call is more reliable
- Does not say what the result is
clawhub · ↓ 2 625 ★ 1