Discover quality open-source projects, submit projects anonymously, and claim and edit your own project.
A little curiosity. A world of open source.
THE FIRST COLLECTIONHeadroom compresses tool outputs, logs, RAG chunks, files, and conversation history before they reach the LLM, reducing token usage while preserving key information. Runs locally as a Python/TypeScript library, proxy, MCP server, or agent wrapper for Claude Code, Codex, Copilot, Cursor, and more.
Caveman is a skill and proxy that compresses the prose of AI coding agents, cutting output tokens by up to 65% and input tokens by about a third. It works with 30+ agents via a simple CLI, offers benchmark data, and includes privacy‑friendly telemetry controls.
