caveman-discover
Find and label every LLM workflow in the repository so Caveman Cloud groups spend by workflow instead of one bucket. Use for "discover workflows" or breaking LLM spend down by workflow.
- installs 8w
- 35,090
- 30-day movement
- starts with the next reading
- Related entries
- 1
- Connections
- 0
caveman-discover is a SKILL.md that guides an AI coding agent through finding and labeling every LLM workflow in a repository, so Caveman Cloud can group spend by workflow instead of a single bucket. It follows five steps: inventory workflows from entry points, name them with an enforced slug grammar, propose a table before applying, verify one labeled path with the repo's own tests or scripts, and report honestly when no LLM entry points exist.
Use it when you want to break LLM spend down by workflow rather than see it as one undifferentiated total in Caveman Cloud.
Use it to
- Discover and label all LLM workflows in a repo
- Break LLM spend down by workflow
- Apply consistent slug names to LLM jobs
- Verify a labeled workflow still passes an existing test or script
For Developers using Caveman Cloud to track LLM spending
- Host repository
- JuliusBrussee/caveman
- Installs, lifetime
- 47k
- Installs, 8 weeks
- 35k
- Host stars
- 106k
- Host language
- Go