An agentic skills framework & software development methodology that works.

A battery-included, POSIX-compatible, generative shell

Registry of agents implementing the Agent Client Protocol (ACP)

ACP adapter for pi coding agent

tmux source code

τ-Bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains

Secure Node.js Execution Without a Sandbox A lightweight library for secure Node.js execution. No containers, no VMs — just npm-compatible sandboxing out of the box.

UlamAI is an open-source Lean theorem prover and formalizer.

TinyAGI is the agent teams orchestrator for One Person Company. (fka TinyClaw)

AI agents running research on single-GPU nanochat training automatically

Harness for running and evaluating AI agents against RL environments

🤗 smolagents: a barebones library for agents that think in code.

Packaging architecture for RL environments

A tool for retrosynthetic planning

Fast, small, and fully autonomous AI personal assistant infrastructure, ANY OS, ANY PLATFORM — deploy anywhere, swap anything 🦀

Harbor is a framework for running agent evaluations and creating and using RL environments.

EvoSkill — An open-source framework that automatically discovers and synthesizes reusable agent skills from failed trajectories to improve coding agent performance on long-horizon tasks.

⚒ Evolutionary self-improvement for Hermes Agent — optimize skills, prompts, and code using DSPy + GEPA

🦞 Just talk to your agent — it learns and EVOLVES 🧬.

Multi-platform SDK for integrating GitHub Copilot Agent into apps and services

CLI-Anything: Making ALL Software Agent-Native

cli for discord with sqlite backend

Memory library for building stateful agents

Visualize PyTorch memory timelines and drill into CUDA allocation call stacks directly from your terminal.

Static security scanner for LLM agents — prompt injection, MCP config auditing, taint analysis. 49 rules mapped to OWASP Agentic Top 10 (2026). Works with LangChain, CrewAI, AutoGen.

Open-source orchestration for zero-human companies

Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents.