agent-skills-eval

Test runner that proves whether an Agent Skill actually improves model output, with a side-by-side report.

Desktopfreeglobal

A desktop test runner that measures whether an Agent Skill improves model responses by running the same prompts twice—with and without the skill—having a judge model grade both, and producing a side-by-side report of the measured lift. It addresses the lack of objective proof before publishing skills and is aimed at developers and teams building or evaluating Agent Skills; it is open-source (MIT), framework-agnostic, and can be self-hosted.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of agent-skills-eval - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim agent-skills-eval to unlock →

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether agent-skills-eval (or something else) actually fits.

Try a problem