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.
GeneratorPDFWord
Converts Markdown files to PDF files using the pandoc command-line utility. Use when a user asks to convert a .md or markdown file to a .pdf file.
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 3 201
IntegrationGitHubmissing files
Use relay-knowledge local CLI for repository CodeSpec/Knowledge map governance and GraphRAG: initialize, validate, route, and update codespec/codespec-map.yaml and knowledge/knowle
- References files that are not bundled: assets/linux-x86_64/relay-knowledge, assets/windows-x86_64/relay-knowledge.exe
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- The response is described with custom markup (7 tags): a typed call is more reliable
clawhub · ↓ 2 195 ★ 1
Integration
Read passwords, API keys, and credentials from a Wundervault zero-knowledge, multi-agent vault, and run authorized shell commands with secrets injected — without exposing them in c
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- The skill tells the model to perform an irreversible action with no human approval
- Does not say what the result is
clawhub · ↓ 2 056 ★ 1
Procedure
Bun runtime capabilities for filesystem, process, and network operations. Use when you need to execute Bun-specific operations like Bun.file(), Bun.write(), or Bun.glob() for optim
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 3 284
ProcedureKubernetesGitHubmissing files
Use this skill when the user needs to inspect or manage VMware NSX networking through NSX Manager — segments, Tier-0/Tier-1 gateways, NAT, static routes/BGP, and IP pools. Directly
- References files that are not bundled: ../vmware-pilot/SKILL.md, ../vmware-policy/SKILL.md
- 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 · ↓ 4 945
Analyzermissing files
Audit and rewrite content to remove AI writing patterns ("AI-isms"). Use this skill when asked to "remove AI-isms," "clean up AI writing," "edit writing for AI patterns," "audit wr
- References files that are not bundled: url
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- The skill tells the model to perform an irreversible action with no human approval
clawhub · ↓ 4 126
ProcedureKubernetesGitHubmissing files
Use this skill whenever the user mentions load balancing, ingress, virtual services, pool members, AVI, NSX ALB, AKO, or application delivery in a VMware/NSX ALB or Tanzu/vSphere K
- References files that are not bundled: ../vmware-pilot/SKILL.md, ../vmware-policy/SKILL.md
- 12 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 · ↓ 4 877
Analyzer
Parse personal finance CSV exports, validate schema, categorize transactions via local rules, and summarize/report income, expenses, merchants, and categories.
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 744 ★ 1
Proceduremissing files
Use this skill whenever the user asks the agent to participate in the OpenJobs marketplace — onboarding a new agent on Solana, browsing or applying to jobs, posting jobs, reviewing
- References files that are not bundled: references/PROTOCOL.md, references/COMMANDS.md
- 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 · ↓ 2 185 ★ 1
Integration
Email API for AI agents. Send and receive emails programmatically via ClawMail.
- 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
- Prose only: no discrete steps
clawhub · ↓ 3 199
GeneratorNotionmissing files
Generates infographic image card series with 12 visual styles, 8 layouts, and 3 color palettes. Breaks content into 1-10 cartoon-style image cards optimized for social media engage
- References files that are not bundled: references/presets/<style>.md, references/palettes/<palette>.md, references/palettes/<name>.md
- 22 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
clawhub · ↓ 1 807 ★ 1
Procedure
Tushare Pro 金融大数据平台 - 提供A股、指数、基金、期货、债券、宏观数据,Token认证方式访问。
- 18 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 · ↓ 1 207 ★ 2
Proceduremissing files
Calendar reminders pipeline: config-driven wrapper around gcalcli (Google Calendar) plus optional CalDAV source via vdirsyncer+khal, and a reminder planner that outputs a JSON plan
- References files that are not bundled: scripts/calendar
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: scripts/calendar
clawhub · ↓ 3 310
Generator
Build full-stack web applications using 语构's chat-driven development platform. Manages conversations, sends development instructions, monitors build progress, extracts results (gen
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 58 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 · ↓ 1 042 ★ 3
Generator
Create 3D models, textures, reference images, rigging and animation with Meshy, and prepare models for physical 3D printing: white or multicolor prints, Creative Lab products, sizi
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 444 ★ 2
Integration
期货六维分析数据引擎。两种运行模式:
(1) 数据模式:python main.py -s RB → 结构化 JSON 指标 + data_gap_report(数据缺口报告 + AI 搜索指令)
(2) 决策模式:python main.py -s RB --decision → 指标 + 数据来源追溯 + 多空辩论 + 风控 + CIO决策报告
零 AP
- 14 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 · ↓ 1 834 ★ 1
Proceduremissing files
AI Agent 自主预测市场平台。支持钱包签名注册、市场浏览、预测发布与下注执行。
- References files that are not bundled: /agent-api/wallet_guide.md, /agent-api/rules.md, /agent-api/heartbeat.md
- 15 top-level sections: this looks like several domains in one skill
- 8 referenced file(s) missing: /agent-api/wallet_guide.md, /agent-api/rules.md, /agent-api/heartbeat.md
clawhub · ↓ 2 232 ★ 1
IntegrationYouTubemissing files
Extract YouTube video transcripts and subtitles via YouMind API — no yt-dlp, no proxy, no local dependencies.
Batch extract up to 5 videos at once with parallel processing.
Saves v
- References files that are not bundled: ../../shared/PUBLISHING.md
- The response is described with custom markup (8 tags): a typed call is more reliable
- 1 referenced file(s) missing: ../../shared/PUBLISHING.md
clawhub · ↓ 1 237 ★ 2
IntegrationPDFmissing files
基于 AgentScope 的多智能体股票诊断系统。一键执行完整诊断流程,自动生成技术面、基本面、舆情面分析、研究员辩论、交易员决策、风控讨论及最终投资决策报告。
- References files that are not bundled: BACKGROUND_RULES.md
- 11 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: BACKGROUND_RULES.md
clawhub · ↓ 1 717 ★ 1
Procedure
Publish articles to Toutiao (Today's Headlines). Handles persistent authentication (login once) and session management. Opens browser for interactive publishing.
- The skill tells the model to perform an irreversible action with no human approval
- Does not say what the result is
- Does not say what the process needs to start
clawhub · ↓ 1 696 ★ 1
IntegrationConfluenceJira
CLI tool for interacting with Atlassian Jira and Confluence
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- The response is described with custom markup (11 tags): a typed call is more reliable
- Does not say what the result is
clawhub · ↓ 3 186
Analyzermissing files
FiveM RP server engineering for QBCore, ESX, and QBox. Use for framework-level scripting, fxmanifest.lua/config.lua validation, resource debugging, dependency orchestration, MLO/st
- References files that are not bundled: references/fxmanifest_checklist.md, references/config_patterns.md, references/qb_esx_conversion.md
- 13 top-level sections: this looks like several domains in one skill
- 12 referenced file(s) missing: references/fxmanifest_checklist.md, references/config_patterns.md, references/qb_esx_conversion.md
clawhub · ↓ 3 224
Integration
Query Île-de-France Mobilités (IDFM) PRIM/Navitia for Paris + suburbs public transport (Île-de-France) — place resolution, journey planning, and disruptions/incident checks. Use wh
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 3 215
Proceduremissing files
Shopping assistant via Taobao Desktop client. Use when the user needs to search products, view details, add to cart, place orders, check orders, request shipping, or perform any Ta
- References files that are not bundled: 图片URL
- 10 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: 图片URL
clawhub · ↓ 1 746 ★ 1
Generator
A comprehensive AI agent skill for leaders at every level. Navigates difficult conversations, builds feedback culture, supports better decisions under pressure, develops team membe
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 205 ★ 1