← All skills

eval-code

100 / 100 · Low risk

Fine to try.

By Calibre-Labs/reforge-ai-evals (GitHub) · coding · 95 stars · scanned at commit 92afb10

Write deterministic, code-based evaluator functions for an AI product. Use this skill whenever you need to write evaluators that check structural properties of AI outputs (format, count, presence, schema compliance) without calling another LLM. Also use when auditing existing code-based evaluators for brittleness, coverage gaps, or false positives.

Works with any agent that supports SKILL.md skills.

What this skill can do

We found nothing in the files we read that matches our rules. That does not prove the skill is safe: see the limits of static scanning.

Install

Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.

if [ -e ~/.agents/skills/'eval-code' ] || [ -L ~/.agents/skills/'eval-code' ]; then echo 'A skill folder named eval-code already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/Calibre-Labs/reforge-ai-evals' "$d" && git -C "$d" checkout --quiet '92afb10f9a692da088134b6d27841f92318aa211' -- 'skills/eval-code' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/eval-code' ~/.agents/skills/'eval-code' && rm -rf "$d"; fi

Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'eval-code' ~/.claude/skills/'eval-code' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.

Badge for your README

The badge for Calibre-Labs/reforge-ai-evals shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/Calibre-Labs/reforge-ai-evals.svg)](https://zalavijaysinh.github.io/skilltrust/repos/Calibre-Labs/reforge-ai-evals/)

Score history