JuliusBrussee/caveman
A Claude Code skill, with matching versions for other agents, that makes the model answer in terse fragments with no filler, while code, commands and error text stay exact. Companion commands write short commit messages and one-line review comments, and another compresses memory files such as CLAUDE.md.
Idea adaptedKey files read, not run
I learnt this repo's idea and adapted it into my own workspace setup.
How we know
We read the README and key files, not every source file, and did not run it, so the most we can say is Idea adapted or Risky.
Good for
Anyone who wants lower output token cost and shorter replies from a coding agent.
Not for
Readers who need full prose explanations, such as docs or onboarding notes.
Repo
- Stars
- 104,650
- License
- NOASSERTIONGitHub could not match this to a standard license. Read the repo's LICENSE file before you use it.
- Maturity
- active
- Last push
- 2026-09-10
- Last checked
- 2026-11-14