gaearongaearonCommunityยท5 Agent Skills Included

woodshed

Test, rate, and refine skill variants automatically

Runs skill variants against test fixtures and scores the results automatically. Eliminates guesswork when comparing prompt versions by showing which variant performs best. Caches past runs so re-testing is instant, and logs every output for easy review. Helps developers iterate on skills faster with clear, repeatable benchmarks.
npx skills add gaearon/woodshed --all -g -y

All Skills in This Repository (5)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install woodshed?โ–ผ

Run `npx skills add gaearon/woodshed --all -g -y` in your terminal to install everything globally.

What does woodshed do?โ–ผ

It runs your skill variants against test fixtures multiple times, then uses an evaluator to rate each result so you can see which version works best.

How do I compare two versions of a skill?โ–ผ

Place each version as a variant folder (like baseline/ and experiment/) under your skill, then run woodshed to get scored results for each one.

Does woodshed re-run old tests every time?โ–ผ

No. It caches past results and skips them on re-runs, unless you pass the --reset flag to force fresh runs.

Is woodshed safe to run on my machine?โ–ผ

It is alpha software that runs Claude in unrestricted mode, which can modify or delete files. Use it in a disposable environment and review the author's warnings first.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’