Agent-Level-Up / caveman-learn
CLAUDE.md3 starsChanged 3 months agoMITView on GitHub
The consent-gated half of caveman learn. The analyzer (the Go proxy) measures where an agent's tokens go and writes a ranked plan; this skill is what an agent loads to act on that plan — proposing each fix and applying it only with the user's per-edit yes. It is the loop-closer the learn spec §10 describes, plus the new cavemem_offload move. See ../../mem/CLAUDE.md (cavemem) · ../caveman-explore/SKILL.md (the packaging precedent)
5 other projects use a version of this.
Adopted by
Most recently taken up by
- mjlueck-hash/cavemanword for word, first seen Nov 8, 2026
- pengawasmknan/claude-skillsword for word, first seen Oct 11, 2026
- nomdedev/agent-manager-templateword for word, first seen Sep 30, 2026
- xingtu1996/xingtu-skillsword for word, first seen Sep 7, 2026
- xingtu1996/dsh-xingtu-skillsword for word, first seen Sep 7, 2026
Other versions5 projects use a version of this.
Word for word · 5
- nomdedev/agent-manager-template★ 2100% same
- xingtu1996/dsh-xingtu-skills★ 2100% same
- mjlueck-hash/caveman★ 1100% same
- pengawasmknan/claude-skills★ 0100% same
- xingtu1996/xingtu-skills★ 0100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- cavecrewSkill
- caveman-commitSkill
- caveman-compressSkill
- caveman-discoverSkill
- caveman-evidence-reviewSkill
- caveman-exploreSkill
- caveman-helpSkill
- caveman-learnSkill
- caveman-manageSkill
- caveman-optimizeSkill
- caveman-reviewSkill
- caveman-setupSkill
- cavemanSkill
- caveman-statsSkill
- fullstack-devSkill
The file
Files
skills/caveman-learn — the Caveman Learn editing skill (MIT, public)
The consent-gated half of caveman learn. The analyzer (the Go proxy) measures
where an agent's tokens go and writes a ranked plan; this skill is what an agent
loads to act on that plan — proposing each fix and applying it only with the
user's per-edit yes. It is the loop-closer the learn spec §10
describes, plus the new cavemem_offload move.
Layout
SKILL.md— the canonical skill body (frontmattername: caveman-learn+ a trigger-phrasedescription; body = the read-plan → per-class consent loop). This file is the source of truth.tests/skill-file.test.mjs— asserts the canonical file is well-formed and honest (frontmatter present; the net-token-negative gate, the never-make-the-agent-dumber guard, consent-per-edit, and reversibility are all stated; no imperative for behavioral findings; no placeholders).
Install path
caveman tools skills install caveman-learn (in ../../cli/src/index.ts) writes this file
into a repo's .claude/skills/caveman-learn/SKILL.md (Claude Code) or
~/.codex/skills/caveman-learn/SKILL.md (Codex). The CLI embeds a byte-identical copy
(CAVEMAN_LEARN_SKILL_MD) because the published CLI ships no sibling assets;
../../cli/tests/skills.runtime.mjs asserts the embedded copy equals this canonical
file (the drift guard). Change this file and that constant together.
Boundary (binding)
The skill — using the agent's own file tools — is the ONLY thing that edits a user's
config. caveman learn apply stays read-only (it materializes candidates), and
caveman mem * are mechanical store ops. The offload move enforces a net-token-negative
gate and the never-make-the-agent-dumber guard before any trim.
See ../../mem/CLAUDE.md (cavemem) · ../caveman-explore/SKILL.md (the packaging precedent)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.