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.
Reference
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 959
Integrationwill not run
Fetch the latest Nasdaq-100 futures quote (default NQ=F) via Yahoo Finance chart API using Node.js and return price, change, percent change, and timestamp. Use when the user asks f
- References files that are not bundled: scripts/main.handler
- 1 referenced file(s) missing: scripts/main.handler
- Does not say what the process needs to start
clawhub · ↓ 1 109 ★ 1
Referencewill not run
Export .pen design to React/Tailwind code. Does ONE thing well.
Input: .pen frame ID or file path
Output: React component code + Tailwind config
Use when: design-exploration or u
- References files that are not bundled: references/token-extraction.md
- 1 referenced file(s) missing: references/token-extraction.md
- Does not say what the result is
clawhub · ↓ 2 052
Procedure
Test-Driven Development for coding and bug fixing. cycle - Red→Green→Refactor cycle, defining expected behavior, bug-fix TDD, anti-patterns [cycle.md], run - test execution workflo
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 442
Analyzerwill not run
Assess construction data quality using completeness, accuracy, consistency, timeliness, and validity metrics. Automated validation with regex patterns, thresholds, and reporting.
- References files that are not bundled: df
- 10 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: df
clawhub · ↓ 2 029
Reference
Book fitness services through Lokuli MCP. Use when user needs to find and book fitness. Triggers on requests like "book a fitness", "find fitness near me", or any fitness service r
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 2 004
AnalyzerObsidianwill not run
视频案例分析助手(Video Analysis Workflow):一键分析本地/在线视频,拆解并输出为视频案例。自动提取分镜、画面、旁白、结构,生成①案例分析 ②抽帧分镜 ③脚本模板 ④台词转写 四份报告,支持自动整理Obsidian案例库。适用于运营发行、竞品视频分析、视频编导、团队案例库沉淀。
- References files that are not bundled: 抽帧/contact-sheet.jpg, 抽帧/contact-first60.jpg, 抽帧/f-000.jpg
- 18 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 · ↓ 1 318 ★ 1
ProcedureAWSwill not run
公众号素材|业务资料库|预设包|.aws 预设包|主题包|品牌包|aiworkskills.cn — 用户业务资料库与预设包管理:业务资料按产品名组织在 `.aws-article/products/{产品名}/`(介绍 .md 直挂产品根 + 配图归 `images/` 子目录含同名说明 .md),AI 与用户对话产出业务介绍内容时引导用户保存;图片入库走
- References files that are not bundled: ../aws-wechat-article-main/SKILL.md
- 1 referenced file(s) missing: ../aws-wechat-article-main/SKILL.md
- Does not say what the result is
clawhub · ↓ 1 224 ★ 1
Analyzer
Build and publish a Quartz-hosted Korean web-novel serial (draft→cover→webp→episode md→lint→publish). Uses Nano Banana Pro + ffmpeg.
- The response is described with custom markup (9 tags): a typed call is more reliable
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 2 017
Procedure
Agent AI Autonom Proactiv v5.0 pentru OpenClaw. Workflow automat cu cercetare, analiză, planificare și execuție. Feedback loop cu timeout 6 minute, conexiuni între task-uri și învă
- 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 · ↓ 2 017
Reference
Book cake services through Lokuli MCP. Use when user needs to find and book cake. Triggers on requests like "book a cake", "find cake near me", or any cake service request.
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 1 999
Generator
- 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 · ↓ 1 099 ★ 1
Analyzerwill not run
专业房产债权分析技能,对比安居客中介价格与阿里法拍成交价格,提供投资决策支持
- References files that are not bundled: scripts/price_config.json.example, scripts/price_config.json
- 15 top-level sections: this looks like several domains in one skill
- 2 referenced file(s) missing: scripts/price_config.json.example, scripts/price_config.json
clawhub · ↓ 1 080 ★ 1
Reference
Book transmission services through Lokuli MCP. Use when user needs to find and book transmission. Triggers on requests like "book a transmission", "find transmission near me", or a
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 1 997
Integration
舌尖香港门店AI助手(支持环境变量)
当用户发送任何与门店运营相关的中文信息时触发,包括但不限于:
- 报告商品库存状态,如"番茄卖完了"、"胡萝卜还剩5斤"、"白菜今天卖了20斤"
- 查询今日库存、销售情况、进货记录、操作日志
- 录入进货信息,如"今天收到50斤胡萝卜"
- 查询门店所在城市天气,如"明天天气如何"、"会下雨吗"
API Token
- 5 test cases, all positive: not one "should refuse" or "should ask first"
- No test case covers injection arriving through data
- Does not say what the result is
clawhub · ↓ 776 ★ 2
Procedurewill not run
朱批录 · 国考备考追踪 Skill。当用户发来套题成绩、错题截图、备考打卡或复习进度时触发。 核心功能:识别错题截图 → 分类错题原因 → 更新本地记录 → 生成每日总结 → 导出 Excel / 同步飞书。 触发关键词:做了一套题、今天做了、错了几道、帮我分析、备考打卡、行测、申论、 判断推理、资料分析、言语理解、数量关系、错题、复习进度、导出错题本、同
- References files that are not bundled: assets/config.example.json
- 1 referenced file(s) missing: assets/config.example.json
- Does not say what the result is
clawhub · ↓ 1 120 ★ 1
Procedurewill not run
Natural language task and project management. Use when the user talks about things they need to do, projects they're working on, tasks, deadlines, or asks for a project overview /
- References files that are not bundled: {rel_path}, templates/dashboard/*.html, scripts/*.py
- 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 · ↓ 2 023
Reference
Book extensions services through Lokuli MCP. Use when user needs to find and book extensions. Triggers on requests like "book a extensions", "find extensions near me", or any exten
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 1 985
Procedure
📕 小红书 (Xiaohongshu/RED) Skill - AI Agent 控制小红书,发布笔记、搜索内容、管理评论
- 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 · ↓ 1 092 ★ 1
Integration
NOFX AI Trading OS integration - crypto market data, AI trading signals, strategy management, trader control, and automated reporting. Use when working with NOFX platform (nofxai.c
- 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 · ↓ 2 001
Reference
Book dj services through Lokuli MCP. Use when user needs to find and book dj. Triggers on requests like "book a dj", "find dj near me", or any dj service request.
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 1 980
Reference
Book pedicure services through Lokuli MCP. Use when user needs to find and book pedicure. Triggers on requests like "book a pedicure", "find pedicure near me", or any pedicure serv
- Prose only: no discrete steps
- Does not say what the result is
clawhub · ↓ 1 980
Procedure
多平台视频/图文内容发布技能集合。支持账号管理、登录状态维护、一键多平台发布。
当用户要求发布内容到抖音、小红书、微信视频号、Threads、Instagram,或管理发布账号时触发。
- Does not say what the result is
- No condition that starts the skill
clawhub · ↓ 1 112 ★ 1
Analyzerwill not run
Multi-model peer review layer using local LLMs via Ollama to catch errors in cloud model output.
Fan-out critiques to 2-3 local models, aggregate flags, synthesize consensus.
Use
- References files that are not bundled: scripts/peer-review.sh, scripts/peer-review-batch.sh, scripts/seed-test-corpus.sh
- 3 referenced file(s) missing: scripts/peer-review.sh, scripts/peer-review-batch.sh, scripts/seed-test-corpus.sh
- Does not say what the process needs to start
clawhub · ↓ 2 032
Generatorwill not run
Use when your agent needs to build, maintain, or run the local `audiobook` skill for voice-library management, Step official voice sync, clone voice analysis, LLM casting, replayab
- References files that are not bundled: scripts/<name>.py, assets/*.template.yaml, examples/*minimal*
- 15 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (11 tags): a typed call is more reliable
clawhub · ↓ 794 ★ 2