skill-autoresearch

What problem does it solve?

Autonomously optimize an existing skill by running it against a test set, scoring outputs with binary evals, and retaining mutations that improve performance.

Core Features & Use Cases

• Automated evaluation loop: run the target skill on a small test set, score outputs with binary evals, and apply targeted mutations that improve results.
• Binary eval framework: leverage the eval-guide to define repeatable, yes/no checks for skill quality.
• Baseline management and mutation logging: keep an untouched baseline and record each experiment for traceability.
• Dashboard and artifacts: generate results.json, results.tsv, and a self-contained status page to monitor progress.
• Reusability: works for any SKILL.md-based workflow that benefits from iterative improvement.

Quick Start

Run autoresearch on a target skill using a small test set to iteratively mutate and improve it.