caveman
A skill for AI coding agents that makes them reply in terse, caveman-style language.
Liked for making coding agents give short, matter-of-fact answers instead of verbose output12345, with token savings seen as a secondary bonus.678 The most common caveat is that its effect is modest or doubtful9101112, and one user says the model starts ignoring it after a while.13
Why people recommend it
Terse responses · 3
Commenters use it to cut LLM verbosity and get short, matter-of-fact replies.
“I use it to make the agent shut up.”
“a coworker told me about Caveman and it has made my Claude experience vastly better.”
Show 1 more
“I recently started using caveman, and it’s been great.”
Token savings · 3
Some use it to save tokens, though others treat the savings as a side benefit.
“In fact, there's even a skill for it: https://github.com/juliusbrussee/caveman”
Show 1 more
“It's for getting it to output shorter answers, but also could help with your burnout.”
Less reading · 2
Users say it cuts the time spent reading and digesting long model output.
“I recently started using caveman, and it’s been great.”
“I use the caveman skill, and I think that does help some.”
Caveats and dissent
Modest effect · 2
Even users who recommend it say it helps only somewhat.
“Ponytail and Caveman seem to help somewhat.”
“I use the caveman skill, and I think that does help some.”
Doubtful value · 2
One commenter calls such token-saving skills snake oil, and another is unsure how useful it still is.
“Whether it's caveman, RTK, or whatever other vibe-coded productivity/token cost saving hacks/skills/claude.md.”
“I don't use any plugins or skill libraries apart from Caveman and I'm not sure how useful that really is anymore”
Gets ignored · 1
One user reports the model stops following the skill after a while.
“I use the caveman skill, but after a while it is just ignored”
Savings less relevant · 2
One user says token efficiency matters less now than when it was released.
“I use it to make the agent shut up.”
Often compared with
Who recommends it in their own words
From the makers not counted in the ranking
“this skill removes all extraneous response text, including articles.”
Threads where it came up
- RTK reports token savings, but our cost benchmarks disagree 3 mentions
- Ask HN: Should I be polite to my agent or should I treat it like shit? 1 mention
- Claude Haiku 5.5 1 mention
- DeepSeek v4.1 Flash 1 mention
- Portal by Spotify cut my Claude Code token usage by 90% 1 mention
- Claudette: Make Claude Stop Talking Like a BuzzFeed Article 1 mention
Also written as Caveman, caveman skill. Data: caveman.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 16
“In fact, there's even a skill for it: https://github.com/juliusbrussee/caveman”
“I don't use any plugins or skill libraries apart from Caveman and I'm not sure how useful that really is anymore”
“I use it to make the agent shut up.”
“Ponytail and Caveman seem to help somewhat.”
“Whether it's caveman, RTK, or whatever other vibe-coded productivity/token cost saving hacks/skills/claude.md.”
“Same. I use this to fix claudespeak.”
“There is also Caveman[1] skill which rewrite prompts.”
“I use the caveman skill, but after a while it is just ignored”
“a coworker told me about Caveman and it has made my Claude experience vastly better.”
“plus maybe "caveman" mode to cull the verbosity”
“I use the caveman skill, and I think that does help some.”
“This is what you want: https://github.com/juliusbrussee/caveman”
“But I still lean towards https://github.com/juliusbrussee/caveman.”
“I recently started using caveman, and it’s been great.”
“It's for getting it to output shorter answers, but also could help with your burnout.”
“I use caveman to help with that”