Caveman is an open-source developer tool designed around a simple yet powerful premise: "why use many token when few do trick." Built by maker Julius Brussee, Caveman allows developers to wrap AI coding assistants and LLM workflows—such as Claude Code, Codex, and Hermes—with a single command.
At its core, Caveman operates as a local proxy that intelligently compresses logs, tool outputs, and source code files before sending them to LLM provider endpoints. In a pinned 54-run benchmark, Caveman reduced input tokens by 33.2% while achieving 18/18 correctness checks, ensuring token savings do not compromise code or task accuracy.
In addition to compressing standard logs and file outputs, Caveman can execute existing agent skills with approximately 70% fewer tokens by rendering and loading text as images. Backed by a thriving open-source ecosystem with over 97,000 GitHub stars, Caveman provides a practical, efficient solution for optimizing developer workflows and reducing API usage costs.
Caveman is a developer tool that wraps AI tools like Claude Code, Codex, and Hermes with a local proxy, compressing logs, tool outputs, and files before provider API calls to minimize token consumption.
In a 54-run benchmark, Caveman achieved a 33.2% reduction in input tokens while passing 18 out of 18 correctness checks. It can also reduce token usage by ~70% when loading text as images for agent skills.
Caveman wraps tools and provider calls including Claude Code, Codex, Hermes, and other agent workflows via its local proxy command.
Yes, Caveman is part of an open-source ecosystem that boasts over 97,000 GitHub stars.
0 community reviews for Caveman
Help the community by sharing your experience. Your review helps others make better decisions.
Can't stop building
