A framework for discovering, compiling, and validating reusable skills for scientific agents.

OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards

Shell and coding agent on mcp clients

Vetix — Automated scanning, identification, and assessment of SKILL security risks.

Build and run agents you can see, understand and trust.

Turn project work into reusable knowledge — an AI-agent skill for Claude Code & Codex

Practical Agent Skills — English-for-engineers coaching, Pi Agent setup, and more. Install via npx skills add.

A Package Manager for the AgentSkills standard. Discover, Use, and Create AgentSkills with Claude Code, Codex, Gemini CLI, OpenCode and more

🦖 Serverless AI Agent Framework with Geo-distributed Edge AI Infra.

Bitbucket MCP - A Model Context Protocol (MCP) server for integrating with Bitbucket Cloud and Server APIs

The LLM Council works together to answer your hardest questions

Writing is thinking! WIT provides skills for scientific thinking and writing!

The container platform tailored for Kubernetes multi-cloud, datacenter, and edge management ⎈ 🖥 ☁️

Learn AI and LLMs from scratch using free resources

The World's First Virtual Terminal for AI Agents

A Python library for building AI agents that leverage the full power of Google Antigravity.

An MCP server for converting Markdown to interactive mind maps with export support (PNG/JPG/SVG).

Toward Skill-Native LLMs: Skill Entropy for Benchmarking and Training Long-Horizon Reasoning

A skill library that teaches Claude Opus the working disciplines of Claude Fable 5. Move the checkpoints, not the capacity.

From thought to skill. From signal to structure.

Web Graphical User Interface (GUI) for Gas Town multi-agent orchestrator - A companion interface for steveyegge/gastown

A curated list of open source GitHub repositories related to ChatGPT, the OpenAI API, and Codex. Searchable via Claude Code skills.

Causal memory layer for AI agents — MCP server that records decision→outcome relationships. Survives compaction.

Autoharness — a self-learning skill layer for Claude Code — distills skills from your real sessions, updates them as you work, and prunes the ones that stop getting used. No daemon, no benchmark.