Tag
#token-optimization
6 open-source projects filed under this tag.
caveman
Caveman is a Claude Code skill that cuts 65% of AI output tokens by making agents talk like caveman — same accuracy, fewer words, lower cost. 70K GitHub stars.
codesight
CodeSight is a universal AI context generator that cuts token usage by 91x across Claude Code, Cursor, Copilot, and Codex. Supports 14 languages, 30+ frameworks, zero dependencies.
headroom
Headroom is a context compression layer for AI agents that reduces token usage by 60-95% while preserving answer quality. Library, proxy, MCP server.
lowfat
Rust CLI that filters noisy shell command output before it reaches AI coding agents, saving up to 91% of tokens with a plugin-based architecture.
opensquilla
Token-efficient microkernel AI agent with local model router, persistent memory, layered sandbox, and support for 20+ LLM providers — same budget, higher intelligence density.
rtk
RTK is a Rust CLI proxy that sits between your AI coding agent and the shell, compressing command output by 60-90% to slash LLM token costs.