agentsclimarketplace

Pre refined skill

Skill barryroodt/refine-skill/e2e/fixtures/pre-refined-skill

Refine an Agent Skill via the skill-forge judge → hitl loop, in a sandboxed Docker container. npx-installable.

Install
npx -y skills add barryroodt/refine-skill --skill pre-refined-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Already-tight skill used to test delta-below-threshold convergence in fewer passes.

SKILL.md

0.4 KB, 61 tokens by cl100k_base, as published. Nobody here has run it

Goal

A pre-refined skill that should score in A range immediately and produce only trade-off-flagged improvements on subsequent passes.

Process

Single-step process. See Anti-patterns for what to avoid.

Anti-patterns

NEVER add unrelated features. NEVER split short lists.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.