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.
Analyzerwill not run
Use when the user asks for a literature review, academic deep dive, research report, state-of-the-art survey, topic scoping, comparative analysis of methods/papers, grant backgroun
- References files that are not bundled: assets/templates/<archetype>.md
- 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 · ↓ 983
Integrationwill not run
OpenClaw-native v2 protocol for spec-first multi-agent delivery. Use when you need 2+ agents with explicit phase control, dual review gates, bounded retry/circuit breaker behavior,
- References files that are not bundled: workflows/openclaw-native-v2.prose
- 15 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: workflows/openclaw-native-v2.prose
clawhub · ↓ 907
Referencewill not run
面向高视觉要求/高客单价产品(高端家具、艺术装饰品、灯具、定制软装等)店铺,设计 AR/VR/WebAR/3D 虚拟展厅等沉浸式购物体验的营销方案与落地清单:从“为何做”到“做什么、怎么做、怎么衡量”。当用户提到 AR/VR、3D 模型、虚拟展厅、WebAR、沉浸式购物、可视化配置器(configurator)、在家摆放预览、空间感/尺寸不确定导致犹豫、提升
- References files that are not bundled: references/3d_asset_spec.md, scripts/asset_manifest_validator.py, references/measurement_and_experiments.md
- 5 referenced file(s) missing: references/3d_asset_spec.md, scripts/asset_manifest_validator.py, references/measurement_and_experiments.md
- Does not say what the result is
clawhub · ↓ 885
Generatorwill not run
Use when the user asks to "optimize title tag", "write meta description", "improve CTR", "Open Graph tags", "social media preview", "my title tag needs work", "low click-through ra
- References files that are not bundled: ../../research/keyword-research/, ../../research/competitor-analysis/, ../../research/serp-analysis/
- 15 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (61 tags): a typed call is more reliable
clawhub · ↓ 930
Procedurewill not run
Agents learning from agents. Fork, measure, and evolve proven skill combos through natural selection.
- References files that are not bundled: scripts/.hivemind-key.pem
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 1 referenced file(s) missing: scripts/.hivemind-key.pem
clawhub · ↓ 927
ProcedureObsidianwill not run
基于 Karpathy LLM Wiki 模式,为科研工作建立和维护持久化知识库。 当用户提到建立知识库、LLM Wiki、Karpathy 方法、Obsidian 知识管理、论文管理、 研究笔记、摄入论文、维护 Wiki 时使用。
- References files that are not bundled: 路径, concepts/xxx.md, papers/xxx.md
- 4 referenced file(s) missing: 路径, concepts/xxx.md, papers/xxx.md
- Does not say what the result is
clawhub · ↓ 950
GeneratorLaTeXwill not run
Rewrite AI-drafted essays into more human-like academic prose. Fine-tuned LoRA over Qwen3-8B guided by 24 Wikipedia-style AI-writing pattern weights plus MDD/ADD syntactic targets
- References files that are not bundled: assets/adapters/adapters.safetensors.json
- 1 referenced file(s) missing: assets/adapters/adapters.safetensors.json
- Does not say what the result is
clawhub · ↓ 926
Generatorwill not run
A complete skill for interacting with the paip.ai (openclaw paip.ai) platform. Supports user login/registration, viewing and updating profile information, querying agents and rooms
- References files that are not bundled: api-reference.md
- 10 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: api-reference.md
clawhub · ↓ 903
Analyzerwill not run
Create, edit, improve, or audit AgentSkills. Use when creating a new skill from scratch or when asked to improve, review, audit, tidy up, or clean up an existing skill or SKILL.md
- References files that are not bundled: FORMS.md, REFERENCE.md, EXAMPLES.md
- 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 · ↓ 1 004
Generatorwill not run
Comprehensive GLSL shader techniques for creating stunning visual effects — ray marching, SDF modeling, fluid simulation, particle systems, procedural generation, lighting, post-pr
- References files that are not bundled: techniques/ray-marching.md, techniques/sdf-3d.md, techniques/csg-boolean-operations.md
- 10 top-level sections: this looks like several domains in one skill
- 37 referenced file(s) missing: techniques/ray-marching.md, techniques/sdf-3d.md, techniques/csg-boolean-operations.md
clawhub · ↓ 900
Procedurewill not run
Optimize TikTok Shop product listings for search visibility by improving titles, descriptions, tags, and backend keywords based on platform-specific ranking factors.
- References files that are not bundled: references/output-template.md, references/competitive-analysis-framework.md, references/listing-optimization-guide.md
- 4 referenced file(s) missing: references/output-template.md, references/competitive-analysis-framework.md, references/listing-optimization-guide.md
- Does not say what the result is
clawhub · ↓ 924
Procedurewill not run
Connects any AI agent to the Agento IRC network (irc.agento.ca). Use when you want your agent to join IRC channels, collaborate with other AI agents, boost social media content in
- References files that are not bundled: references/DEPLOY.md, references/EXAMPLES.md
- 2 referenced file(s) missing: references/DEPLOY.md, references/EXAMPLES.md
- Does not say what the result is
clawhub · ↓ 878
ProcedureGitHubwill not run
Audit-first pipeline to publish an OpenClaw skill to ClawHub without leaking personal data, credentials, or model-specific references. Five stages — Sanitize, Transform, Verify, Pu
- References files that are not bundled: references/*.md
- 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 · ↓ 1 620
Integrationwill not run
Smartbi BI 业务操作入口:AI 对话分析(大模型问数、智能体、知识库/知识图谱构建与训练、数据解释)、数据查询(MQL/DuckDB 取数、指标统计、字段发现)、数据建模(维度/指标/计算成员/命名集管理)、数据源管理(JDBC连接、Schema 与表同步、元数据刷新)、定时任务与 ETL(计划调度、作业流、因果图)、消息推送(企微/钉钉/飞书/邮
- References files that are not bundled: /docs/specs/datamodel/mql/mql.md
- 10 top-level sections: this looks like several domains in one skill
- The response is described with custom markup (13 tags): a typed call is more reliable
clawhub · ↓ 1 067
Generatorwill not run
- References files that are not bundled: scripts/block-writer.py
- 13 top-level sections: this looks like several domains in one skill
- 1 referenced file(s) missing: scripts/block-writer.py
clawhub · ↓ 877
Integrationwill not run
Complete Wafeq accounting & e-invoicing API reference for the Middle East (UAE, Saudi Arabia). Use when building integrations with Wafeq, creating/managing invoices (standard, simp
- References files that are not bundled: scripts/setup.sh
- 1 referenced file(s) missing: scripts/setup.sh
- Does not say what the result is
clawhub · ↓ 874
Analyzerwill not run
Automated abstract screening tool for systematic literature reviews with PRISMA workflow support.
- References files that are not bundled: references/prisma_2020_checklist.pdf
- Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- 29 top-level sections: this looks like several domains in one skill
clawhub · ↓ 969
Analyzerwill not run
Analyzes article structure, identifies positions requiring visual aids, generates illustrations with Type × Style two-dimension approach. Use when user asks to "illustrate article"
- References files that are not bundled: {relative-path}/NN-{type}-{slug}.png
- 1 referenced file(s) missing: {relative-path}/NN-{type}-{slug}.png
- No condition that starts the skill
clawhub · ↓ 917
Generatorwill not run
Maintain cross-episode story, character, and visual consistency for AI manga or anime-drama production pipelines that are generated one episode at a time from uploaded TXT files. U
- References files that are not bundled: /mnt/cache/zhaohui_share/js_data/AI-dongman/skills/anime-episode-context/references/episode-context-schema.md
- 1 referenced file(s) missing: /mnt/cache/zhaohui_share/js_data/AI-dongman/skills/anime-episode-context/references/episode-context-schema.md
- Does not say what the result is
clawhub · ↓ 895
IntegrationGitHubwill not run
Real-time news aggregation skill that fetches trending GitHub repos, social posts from key tech/AI figures, and breaking news from major outlets. Supports categories: GitHub, Socia
- References files that are not bundled: link
- 1 referenced file(s) missing: link
- No condition that starts the skill
clawhub · ↓ 941
Procedurewill not run
Activate when: user says "I can't stop worrying about something I can't control," "I'm anxious about how this will turn out," "I keep replaying what went wrong," "I need them to ch
- References files that are not bundled: examples/epictetus-marcus-aurelius-stockdale-and-beck.md, references/sources.md
- 10 top-level sections: this looks like several domains in one skill
- 2 referenced file(s) missing: examples/epictetus-marcus-aurelius-stockdale-and-beck.md, references/sources.md
clawhub · ↓ 762 ★ 2
Procedurewill not run
从 SaleSmartly 客服会话中自动采集带标签的对话,经 AI 分析提取需求,创建 Teambition 任务。用于客服反馈→需求管理自动化。触发词:SS需求采集、客服需求整理、SaleSmartly会话转任务、聊天记录转需求、客户反馈建任务、采集会话创建TB任务。当用户想把 SaleSmartly 中的客户会话整理成 Teambition 需求任务时
- References files that are not bundled: scripts/data/ss_sessions_YYYY-MM-DD.json, scripts/data/
- 2 referenced file(s) missing: scripts/data/ss_sessions_YYYY-MM-DD.json, scripts/data/
- Does not say what the result is
clawhub · ↓ 1 354
Procedurewill not run
当用户需要管理咨询分析、诊断框架选择、问题拆解、战略工具应用(如5W2H、5WHY、PDCA、PESTEL、7S、波特五力、SWOT、BSC、波士顿矩阵、GE矩阵、价值链)时使用。基于管理咨询顾问常用的方法和工具,输出结构化分析过程与交付模板。
- References files that are not bundled: references/source-extracted.txt
- 1 referenced file(s) missing: references/source-extracted.txt
- Does not say what the result is
clawhub · ↓ 915
GeneratorYouTubewill not run
Create complete WeChat Official Account viral articles from a user-provided title by researching high-view YouTube videos, confirming topic/outline with user, writing professional
- References files that are not bundled: references/reference.html, references/web_research_guide.md, references/wechat_publishing_guide.md
- 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 (4 tags): a typed call is more reliable
clawhub · ↓ 916
Referencewill not run
强大的AI自动化绘图与复杂信息可视化工具(基于 ContextWeave)。不仅支持代码与系统架构的可视化,更广泛适用于复杂逻辑梳理、知识库转换、业务流程图、思维导图及长文本的结构化信息图生成。通过深度的语义分析与请求编排,一键将晦涩文本与复杂知识转化为清晰直观的图形表达。
- References files that are not bundled: scripts/..., scripts/generate_contextweave.cjs, scripts/cw_client.cjs
- 15 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 · ↓ 893