Skills Directory

Single 70-line CLAUDE.md derived from Karpathy's LLM coding-pitfall observations — #1 GitHub trending for 28 consecutive days.

A complete software development methodology and composable skills framework for coding agents — brainstorm, plan, execute, review, ship.

Agent skill that makes your coding agent think like the laziest senior dev in the room — skip unnecessary code, reuse existing patterns, and write 54% less code while staying 100% safe.

Production-grade engineering skills for AI coding agents — addyosmani's curated bundle covering Claude Code, Cursor, and Antigravity workflows.

Turn any codebase into a queryable knowledge graph — deterministic AST parsing via tree-sitter, Leiden clustering, no vector store required.

ComposioHQ's curated collection of 100+ production-ready Claude Code skills and plugins organized across 10 categories.

Curated DESIGN.md files extracted from real brand design systems so AI coding agents generate UI that matches a specific visual identity.

Turn any website into clean, LLM-ready data — web scraping API built for AI applications.

Connect AI agents to browser control — click, fill forms, scrape, and interact with any website using GPT-4, Claude, or Gemini.

100+ ready-to-use Claude Code agents, commands, hooks, and integrations — installable via CLI or interactive dashboard at aitmpl.com.

Official Anthropic-maintained collection of 20+ MCP server implementations for filesystem, GitHub, Slack, databases, and more.

Multi-source research skill that aggregates and synthesizes content from the last 30 days across platforms — turning scattered signals into structured intelligence.

Open-source, LLM-friendly web crawler that extracts clean data optimized for AI consumption.

ByteDance's research superagent — long-horizon research with web search, coding, and report generation.

Agent skill that removes signs of AI-generated writing from text — rewrites agent output so it reads like a person wrote it, matching voice and preserving facts.

Universal memory layer for AI agents — remembers user preferences, past interactions, and context across sessions.

Vectorless, reasoning-based RAG — PageIndex builds a hierarchical tree index over long documents and uses LLM reasoning to retrieve, instead of cosine similarity over embeddings.

Injects up-to-date, version-specific documentation into LLM context — eliminates hallucination from stale training data.

Official Chrome DevTools MCP server — AI agents can debug, profile, inspect DOM, and analyze web performance.

1000+ pre-built tool integrations with managed OAuth for AI agents — Gmail, Slack, GitHub, Notion, and more.

Build AI copilot experiences in your React app — in-app chat, AI-powered forms, and generative UI.

39 editorial diagram types for Claude Code, Codex, and Pi — self-contained HTML and SVG output with no shadows, no Mermaid slop, and brand-customizable styling.

Anti-AI-slop design skill — 57 slop-test gates, 20 themes, and a self-critique layer that force Claude Code, Cursor, and Codex to generate UI that looks intentionally designed, not template-stamped.

Adaptive web scraping framework that handles anti-bot measures, CAPTCHAs, and dynamic content.

ByteDance's open-source multimodal desktop agent — sees your screen and operates your computer.

Academic-vertical skill bundle for Claude Code and Cursor — literature review structuring, citation-graph traversal, methodology critique, and peer-review draft templates.

AI-powered code review that catches bugs, security issues, and style violations before they reach production.

GitHub's official MCP server — interact with repos, issues, PRs, code search, and notifications via AI agents.

Autonomous research agent that searches the web, synthesizes findings, and produces cited research reports.

Output-style skill that reshapes every coding agent reply to be action-first, numbered, and free of preamble — designed for ADHD-friendly workflows but useful for anyone who wants concise responses.

Matt Pocock's personal Claude Code skill directory — 21 dev workflow skills for TDD, architecture, planning, and git safety, installable via npx.

Scientific-vertical skill bundle for Claude Code, Cursor, and Codex — chained skills covering protein-structure prediction, literature scraping, citation-graph traversal, and methodology comparison.

Open-source automation platform with 400+ integrations and native MCP server support.

The fast, Pythonic way to build MCP servers — like FastAPI but for the Model Context Protocol.

Text-to-image generation with fine-grained style control, consistent characters, and batch production.

Open-source LLM observability — trace, evaluate, and debug your AI agent workflows in production.

Semantic code retrieval and editing MCP toolkit — the IDE for your AI agent with symbol-level navigation.

AI browser automation framework from Browserbase — natural language web interactions with built-in reliability.

Build n8n automation workflows via natural language — AI agents create, edit, and manage workflows through MCP.

Agent skill that turns codebases and system descriptions into interactive, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams — self-contained HTML with motion and crisp export.

Automatically generate comprehensive unit tests from source code with edge case coverage and mocking strategies.

AI reads Twitter/X, Reddit, YouTube, and GitHub — multi-platform social intelligence for agents.

Semantic search across entire codebases — find code by what it does, not just what it's called.

Desktop sandboxes for computer-use agents — secure, isolated environments for AI to operate desktops.

MCP server that feeds Figma design context to AI coding agents for pixel-perfect design-to-code workflows.

Google's MCP toolbox for databases — connect AI agents to PostgreSQL, MySQL, BigQuery, Spanner, and more.

Secure cloud sandboxes for AI agents to execute code — isolated environments with built-in safety.

Automatically expose FastAPI endpoints as MCP tools — turn any API into an AI-callable tool in one line.

Chrome extension MCP server — AI agents control your browser directly from the extension.

Full podcast transcription with speaker diarization, timestamps, and exportable show notes.

Deep web research with multi-source synthesis, citation tracking, and structured report generation.

817 structured cybersecurity skills for AI agents — mapped to MITRE ATT&CK, NIST CSF 2.0, MITRE ATLAS, D3FEND, and NIST AI RMF.

Official AWS MCP servers — AI agents interact with S3, Lambda, EC2, CloudFormation, Bedrock, and more.

React/TypeScript library for building AI chat interfaces — customizable, accessible, production-ready.

AI-driven ETL pipelines that transform, clean, and load data with intelligent schema mapping and error handling.

Security vulnerability scanning for infrastructure with compliance checking and remediation guidance.

Build real-time voice AI agents — speech-to-speech conversations with sub-second latency.

Full-stack MCP application framework — build complete AI apps with MCP servers, clients, and UIs.

Automated code refactoring with safety checks, test verification, and incremental migration strategies.

Transcribe and summarize YouTube videos with timestamps, key points, and chapter-based navigation.

220+ Claude Code skills and agent plugins covering engineering, marketing, product, compliance, and more.

Reverse engineering via MCP — AI agents can decompile, analyze, and understand binary code using Ghidra.

Generate meaningful commit messages and PR descriptions from diffs using conventional commit standards.

Remote MCP server for any GitHub repository — agents can query real code without hallucination, zero setup.

150+ cybersecurity tools accessible via MCP — pentesting, vulnerability scanning, OSINT, and more.

Create multi-step automated workflows connecting 200+ apps with natural language instructions.

Universal SEO skill for Claude Code — 25 sub-skills and 18 sub-agents covering technical SEO, E-E-A-T, schema, GEO/AEO, backlinks, local SEO, maps intelligence, and reporting.

Parse application logs with AI to surface anomalies, error patterns, and root cause analysis.

Deep research agent powered by Firecrawl — produces comprehensive, multi-source research reports.

Agent monitoring, cost tracking, and evaluation — observability built specifically for AI agents.

ComposioHQ's curated directory of practical OpenAI Codex skills for automating workflows across the Codex CLI and API.

Self-organizing AI second brain for Obsidian + Claude Code — drop any source and Claude reads, links, and files it into a connected knowledge graph of plain Markdown you own.

Draft, categorize, and auto-respond to emails with context-aware tone matching and priority sorting.

AI-native web search via Exa — cleaner, more relevant results than Google, optimized for LLM consumption.

Meta-skill that designs domain-specific agent teams — generates specialist agents, orchestrator skills, and team architectures from six battle-tested patterns.

Marketplace of 100+ product management agentic skills — covers roadmapping, PRDs, user research synthesis, prioritization frameworks, and stakeholder communication.

NVIDIA's security scanner for AI agent skills — detects vulnerabilities, malicious patterns, and security risks across 64 vulnerability patterns in 16 categories.

TypeScript-typed skill registry for Claude Code, Cursor, Antigravity IDE, and Copilot — cross-harness skills with strict schema validation.

The open agent skills framework from Vercel Labs — install Claude Code skills, MCP servers, and agent tools with a single npx command.

Smart CI/CD pipeline management with rollback intelligence, canary deployments, and deployment confidence scoring.

Cloud cost analysis with AI-driven recommendations for right-sizing, reserved instances, and waste elimination.

Google's official agent skills for Cloud, Gemini API, and Google products — 13 skills covering BigQuery, GKE, Cloud Run, Firebase, security, reliability, and cost optimization.

Automated incident triage, runbook execution, and team coordination with escalation intelligence.

Comprehensive on-page SEO audit with actionable recommendations, structured data validation, and competitor comparison.

Official CrewAI tools library — 15+ pre-built tools for search, file ops, web scraping, and more.

Monitor trending topics across social media, news, GitHub, and forums with intelligent alerts and analysis.

Track and analyze competitor products, features, pricing, and market positioning with automated monitoring.

Cross-reference claims against multiple authoritative sources with confidence scoring and citation links.

AI-powered meeting scheduling that understands preferences, time zones, and priority to find the perfect time.

AI-assisted video editing with scene detection, auto-cuts, transitions, and caption generation.

Optimize existing content for search rankings with NLP-driven recommendations and SERP analysis.

Cross-platform social media posting with scheduling, content adaptation, and engagement analytics.

Access 3000+ Apify web scrapers and automation actors through MCP — enterprise-grade data extraction.

Parse, summarize, and extract key findings from academic papers on ArXiv with structured output.

Deno's open-source security firewall for AI agents — interposes between agents and production systems, parsing wire-level protocols and enforcing action-level rules in HCL.

Automated web form completion using structured data and intelligent field mapping with browser automation.

Google's Agent Skills library for the Stitch MCP server — design-to-code conversion following the open Agent Skills standard, compatible with Antigravity, Gemini CLI, Claude Code, and Cursor.

Generate click-worthy video thumbnails with AI-optimized text placement, color contrast, and emotion analysis.

GitHub's official awesome list for Copilot — community-curated skills, prompts, and custom instructions for GitHub Copilot users.

AI-powered keyword discovery with search volume estimation, clustering, and content gap analysis.

Discover high-quality link-building opportunities through competitor analysis and content gap identification.

Learning-vertical skill bundle for Claude Code, Cursor, and Codex — curriculum design, retrieval practice, spaced-repetition prompts, and worked-example generators.

Open-source inference server that profiles your hardware, picks the best local model, and plugs into the coding agent you already use — private, offline, zero API costs.

Pipeline skill that ingests URLs, PDFs, transcripts, and arbitrary docs into a NotebookLM-ready format for batch podcast and digest generation.

Operator how-to guides packaged as a Claude skill bundle — practical playbooks distributed via the .claude/skills/ convention.

Team-level skill composition system by Bolt where one prompt triggers an entire stack of skills across your team's workflow.

Vision bridge for text-only coding agents — paste an image into chat and get structured JSON evidence (OCR, layout, semantics) without saving files or switching models.