← All skills

benchmarking-the-agent

82 / 100 · Low risk

Fine to try. Some of what it can do is expected: a coding skill usually runs build and test commands.

By amd/gaia (GitHub) · coding · 1,622 stars · scanned at commit 3ec6ceb

Measure the flagship GAIA agent against Claude Code and across models — quality, truthfulness, steps, tokens, time and real cost — with `gaia eval tasks`. Use when asked to benchmark the agent, compare models or harnesses, check whether a change cost quality, work out what a run costs, or reproduce the harness × model table.

Works with any agent that supports SKILL.md skills.

What this skill can do

Install

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

if [ -e ~/.agents/skills/'benchmarking-the-agent' ] || [ -L ~/.agents/skills/'benchmarking-the-agent' ]; then echo 'A skill folder named benchmarking-the-agent already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/amd/gaia' "$d" && git -C "$d" checkout --quiet '3ec6ceb6429b99c4af63250689239478a4b06c34' -- '.claude/skills/benchmarking-the-agent' && mkdir -p ~/.agents/skills && cp -R "$d"/'.claude/skills/benchmarking-the-agent' ~/.agents/skills/'benchmarking-the-agent' && 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/'benchmarking-the-agent' ~/.claude/skills/'benchmarking-the-agent' (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 amd/gaia shows the score of its lowest-scoring skill.

[![SkillTrust](https://zalavijaysinh.github.io/skilltrust/badges/amd/gaia.svg)](https://zalavijaysinh.github.io/skilltrust/repos/amd/gaia/)

Score history