Connect your devices into a secure WireGuard®-based overlay network with SSO, MFA and granular access controls.

Intercept any app, then call it from Python like a library

A native macOS UI for Apple's platform

Open, agent-agnostic MCP server exposing macOS computer use to any MCP client (Claude Code, Cursor, Codex, …). Background-safe, works on every app.

Native iPhone app for your Hermes agent

Let Claude (or any LLM) actually watch a video — scene-aware, deduplicated frames + transcript, from a URL or local file. Runs locally, MIT.

Work in progress! I don't recommend looking at the code right now.

An interactive RLM (Recursive Language Model) agent built on DSPy.

Open reimplementation of Sakana Fugu — the 'one model to command them all' LLM orchestrator. Read → run → train → serve.

Self-contained UV scripts for data & ML tasks — OCR, vision, audio & more — run one in a command, locally or on Hugging Face Jobs. Built for humans and agents.

Self-hosted instant sandboxes for coding agents on Cloudflare.

Pi coding-agent extension for pruning tool-call trees

tmux-style PTY session multiplexer as a Zig package. Long-lived daemon owns sessions and PTY child processes; clients attach over UNIX-domain JSON-RPC.

Drop-in replacement for `claude -p` that drives the interactive Claude Code TUI inside an in-process tmux PTY session. Go version.

An experimental pi extension that runs and manages qwen with llama.cpp

Lightpanda: the headless browser designed for AI and automation

DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm

Review-first terminal diff viewer for agentic coders

📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG

Pi extension that globally hides skills from the model while preserving /skill:name.

The sandbox agent framework.

Headless CLI client for stateful Agent Client Protocol (ACP) sessions

macOS menu bar app that tells you, in plain English, what each USB-C cable plugged into your Mac can actually do

Plano is an AI-native proxy server and data plane for agentic apps. Smart LLM routing, observability, agent orchestration, and guardrails so you stay focused on your agents core logic.

🤗 ml-intern: an open-source ML engineer that reads papers, trains models, and ships ML models

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills.

A benchmark for evaluating AI agents on realistic business workflows