Caveman Discover

Caveman-discover walks a repo from its entry points (HTTP handlers, cron jobs, CLI scripts, eval harnesses, agent graphs) and proposes a table of LLM workflows, each named by the job it performs rather than the tech behind it. You'd reach for it when all your gateway spend lands in one `unlabeled-workflow` bucket and you want per-job cost breakdowns. The useful part is the discipline: it shows you the labeling table before touching code, wires labels at the callers instead of shared helpers, and flags anything it had to guess from a filename as "review." The one thing to watch is that slugs are close to permanent, since renaming later splits your spend history, so settle the names before you apply them.

skills.sh installs: 119K
GitHub stars: 118.7K
last updated: Oct 26, 2026

View on GitHub

Install

$ npx -y skills add juliusbrussee/caveman --skill caveman-discover --agent claude-code

Installs into .claude/skills of the current project.

Files

You are labeling this repository's LLM workflows for Caveman Cloud. A workflow is a job the code performs — "answer a support ticket", "build the nightly digest", "run the eval suite" — not a technology. Every gateway request can carry a workflow label; unlabeled traffic all lands in one unlabeled-workflow bucket. Your job: find the workflows, name them well, wire the labels, and verify nothing broke.

This changes code, so it goes through the user's normal review: propose the table first, apply after the user agrees. Re-running on an already-labeled repo must change nothing (idempotent).

This skill is operator-invoked. An unlabeled-traffic Cave Plan observation is review-only and does not create an advisory file, proposal, or Draft PR. Do not infer that telemetry selected a callsite or authorized an edit. Independently inventory the repository, present the labeling table, and wait for the user's approval before changing code.