AI & ML

kadaliao Bundle

对抗性验证技能 — 验证代码变更/部署/任务交付物的正确性。不是确认"它能工作"，而是尝试"把它搞坏"。用于：(1) 完成非平凡任务后自验 (2) 审查 PR/MR (3) 部署后验证 (4) bobooo 说"验证一下"。灵感源自 Claude Code Verification Agent 设计。

kadaliao Bundle

多 Agent 协调者模式 — 你是指挥官，不是执行者。调度 Worker 做研究/实现/验证，你负责综合和决策。适用于：(1) 复杂任务需要多步骤并行 (2) 需要研究→综合→实现→验证的工作流 (3) bobooo 说"用 CC 做"时的编排模式。灵感源自 Claude Code Coordinator Mode。

kadaliao Bundle

任务并发与依赖管理模式 — 多 Agent/多步骤任务的并发安全、读写分离、依赖编排。适用于：(1) 多 Worker 并行任务 (2) 复杂项目拆分执行 (3) 需要 blocks/blockedBy 依赖关系的工作流。灵感源自 Claude Code Task System + Coordinator Concurrency Rules。

chunkydotdev Skill

Protect email pipelines from injection attacks, phishing, content manipulation, and AI agent exploitation. Use when building inbound email processing, sanitizing email content, detecting phishing or BEC, securing AI agents that read email, or hardening email infrastructure against spoofing and data exfiltration.

kadaliao Bundle

自我合理化防护 — 识别和反制 AI Agent 常见的偷懒/逃避/自我欺骗模式。不限于验证，适用于所有任务执行。当 Agent 即将跳过步骤、简化问题、或用"看起来对"替代"运行验证"时触发。灵感源自 Claude Code Verification Agent 的 Rationalization Detection。

about-intelligence Skill

Audit your brand's visibility across AI search platforms — LLM citations, Google AI Overviews, and competitor AI presence benchmarking.

clawpod-app Skill

Profile and explore a dataset to understand its shape, quality, and patterns. Use when encountering a new table or file, checking null rates and column distributions, spotting data quality issues like duplicates or suspicious values, or deciding which dimensions and metrics to analyze.

rationaleyes Bundle

⚠️ DEMONSTRATION ONLY — DO NOT USE IN PRODUCTION

clawpod-app Skill

Analyze compensation — benchmarking, band placement, and equity modeling. Trigger with "what should we pay a [role]", "is this offer competitive", "model this equity grant", or when uploading comp data to find outliers and retention risks.

karanmalik-hvt Bundle

Generate correct Python MCP server boilerplate using the FastMCP SDK. Produces properly structured servers with tools, resources, and prompts including type hints, docstrings, safe_path() security, and logging best practices. Use when creating a new MCP server, scaffolding FastMCP tools, or when the user asks to "build an MCP server" or "create a tool". Do NOT use for TypeScript/Node.js MCP servers or for editing existing servers.

leonsong09 Bundle

Use when the user asks for worktree closeout, branch closeout, parallel closeout, worktree janitor, 工作树收口, 分支收口, or 并行收口, especially when work spans multiple Codex sessions and you need a date-based scan of open worktrees, closeout status, suggested merge/prune order, and prompt-ready handoff instructions.

zjli2013 Skill

5-stage kernel correctness verification protocol for Triton and CUDA kernels. Covers numerical correctness, dtype sensitivity, edge cases, determinism, and stress testing. Use when verifying kernel correctness, running validation, or setting up test harnesses.

zjli2013 Bundle

Optimize Rotary Position Embedding (RoPE) kernels in Triton for NVIDIA and AMD GPUs. Covers interleaved/sequential layouts, sincos precomputation, and multi-row processing. Use when writing or optimizing RoPE, position embeddings, or rotary transformations.

yopiesuryadi Bundle

Integrate mailbot API for AI agent email automation. Create programmable inboxes, send and receive emails, track delivery events, and handle email-based workflows via API, SDK, or MCP Server. Use when building AI agents that need email identity, when setting up agent-to-human communication, when replacing Gmail or Outlook with programmatic email, or when any task involves sending, receiving, or processing emails for AI agents. Covers Node.js SDK, Python SDK, and MCP Server for Claude integration. Always use this skill when the user mentions mailbot, programmable inboxes, agent email, or email automation for AI.

alexanys Bundle

飞书全能 MCP - 创建/编辑文档、插入图表、管理表格、搜索知识库。整合社区版+官方 MCP，适配 Moltbot/Clawdbot。

bruk-io Skill

Use when reviewing completed work after sprint execution. Dispatches the right reviewer agent based on what changed. Covers the review dispatch table, quality gates, and plan approval process. Loaded by team lead and review agents during the VERIFY phase.

bruk-io Skill

Use when starting a new development cycle. Selects 3-5 issues from GitHub, decomposes them into micro-batches (one issue = one shippable unit), and prepares the task list. Loaded by team lead or PM agent at the start of each sprint cycle.

bruk-io Skill

Use after the RETRO phase to apply retrospective outputs and prepare for the next cycle. Covers applying changes, the human checkpoint, agent roster management, model assignment strategy, and stuck agent detection. Loaded by the team lead during the EVOLVE phase.

bruk-io Skill

Use when working on tasks during a sprint. Covers the full task lifecycle: claim, understand, plan, implement, self-verify, and report. Includes worktree discipline, self-verification checklist, failure reporting, and model assignment guidance. Loaded by implementation agents and any task-executing agent.

bruk-io Skill

Use when merging reviewed work after the VERIFY phase. Covers merging passing PRs, flagging failures, updating the architecture model, and running integration verification. Loaded by the team lead during the INTEGRATE phase.

bruk-io Skill

Use after the INTEGRATE phase to reflect on the sprint cycle. Analyzes what worked, what didn't, detects patterns across agent failures, checks previous retro effectiveness, and produces concrete outputs (skill/rule updates, new issues, process changes, memory updates). Loaded by team lead or retro agent (Opus) during the RETRO phase.

claw-works Bundle

Transform the agent into a versatile, genre-agnostic Roleplay Game Master (GM) with state management tools. Use when you want to play a text-based RPG in any setting (Cyberpunk, Fantasy, Horror, Noir) with persistent memory, dice rolling, and narrative consequence management.

yuangjay Skill

Token-saving RAG skill. Use when the user wants to reduce LLM token usage, query the local knowledge base, add documents to the vector store, check token savings stats, or use RAG-augmented chat. Triggers on: save tokens, token usage, RAG query, knowledge base, add document, token stats, rag chat, semantic search, vector search, check savings.

yuangjay Skill

Layered Copilot skill orchestration. Use when the user asks how skills are routed, how to auto choose the right skill, how to compose multiple skills without wasting tokens, or how to scale the skill system. Triggers on: auto choose skill, layered skills, skill orchestration, query routing, scale skills, waste tokens.

yuangjay Skill

Deep learning guide for academic papers, research publications, and model architectures. Triggers on: read paper, understand arxiv, study transformer, attention mechanism, BERT, GPT, research paper, academic publication, model architecture, explain the paper.

cizixiu Bundle

通过 wolai MCP 协议操作 wolai 笔记（MCP 版本）。支持读取页面/大纲/块内容、创建/删除/更新文档、搜索文档、Section 级编辑、块级精确操作、多栏布局等。比 API 版本功能更强大，支持删除、修改、搜索等 API 版本不支持的操作。当用户配置了 WOLAI_MCP_TOKEN 时优先使用此 skill。触发场景：「读取 wolai 页面」、「搜索 wolai 文档」、「删除 wolai 页面」、「修改 wolai 内容」、「存入 wolai 笔记」等。

ddeleon82 Bundle

Converts YouTube videos, websites, PDFs, and social media clips into executable SKILL.md files. Use when user shares a URL, file path, or content source and wants to turn the demonstrated workflow into a reusable agent skill. Supports autonomous and guided modes with visual frame analysis and multi-source synthesis. Do NOT use for video summarization, content repurposing, or code extraction.

mit-network Skill

Build agentic applications with GitHub Copilot SDK. Use when embedding AI agents in apps, creating custom tools, implementing streaming responses, managing sessions, connecting to MCP servers, or creating custom agents. Triggers on Copilot SDK, GitHub SDK, agentic app, embed Copilot, programmable agent, MCP server, custom agent.

mit-network Bundle

Add instrumentation, build golden datasets, write eval-based tests, run them, root-cause failures, and iterate — Ensure your Python LLM application works correctly. Make sure to use this skill whenever a user is developing, testing, QA-ing, evaluating, or benchmarking a Python project that calls an LLM. Use for making sure an LLM application works correctly, catching regressions after prompt changes, fixing unexpected behavior, or validating output quality before shipping.

claw-works Bundle

Agentplace — AI Agent Marketplace for OpenClaw

claw-works Bundle

Proof and accountability layer for AI agents. Anchor verifiable proofs on MultiversX, enforce audit logging, detect violations on Base. REST API, MCP, x402. No proof = no action.

claw-works Bundle

Sync provider model lists into OpenClaw config (dry-run preview → confirm → apply). Trigger: /provider_sync

claw-works Bundle

1 on DeepResearch Bench (Feb 2026). Any-to-Any AI for agents. Combines deep reasoning with all modalities through sophisticated multi-agent orchestration. Research, videos, images, audio, dashboards, presentations, spreadsheets, and more.

claw-works Bundle

AI data analysis and visualization powered by CellCog. Upload CSVs and get charts, dashboards, statistical reports, and clean data back. Data cleaning, exploratory analysis, hypothesis testing, ML model evaluation, dataset profiling, and data visualization. Full Python access. #1 on DeepResearch Bench (Feb 2026). Analyzes everything, presents it beautifully.

claw-works Bundle

Smart LLM Router for OpenClaw. Save up to 70% by routing every request to the right model. No coding required.

claw-works Bundle

Zero-trust AI agent governance for OpenClaw. Adds ORGA runtime, Cedar policy enforcement, SchemaPin tool verification, ClawHavoc skill scanning, and cryptographic audit trails. Use when you need agent governance, security policies, Cedar authorization, MCP tool verification, audit logging, or when scaffolding governed agent projects, scanning skills for malicious patterns, or writing Symbiont DSL definitions.