gaia-testing

79 / 100 · Review before installing

Check before installing.

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

GAIA's multi-tier regression harness — runs unit, integration, and real-world (on-machine) tiers and brings back screenshots, logs, traces, planted-fact retrieval proof, and per-operation timing with anomaly flags, all collected to the local machine (no remote login needed to view). It carries GAIA's drivers (Agent UI, Agent UI MCP), exact commands, ports, and `gaia eval agent` baselines, and extends (does not replace) the generic `testing` skill. Fires when the user wants real-world / on-hardw…

Works with any agent that supports SKILL.md skills.

What this skill can do

Runs commands on your computer

This is disclosed by the author in the skill's SKILL.md or README.

Found in 6 places.

.claude/skills/gaia-testing/SKILL.md, line 100git stash && python -m pytest <failing-subset> 2>&1 | tee <run-dir>/unit_base.txt; git stash pop
.claude/skills/gaia-testing/SKILL.md, line 114gaia eval agent --category <cat> # prints an absolute path, e.g. Output: /…/gaia/eval/results/<run-id>/
.claude/skills/gaia-testing/SKILL.md, line 183python util/machine_turn.py --label "#<issue> repro" -- <command>

Show 3 more

.claude/skills/gaia-testing/SKILL.md, line 205ps aux | grep -E "lemonade|llama-server|gaia.ui.server" | grep -v grep # only pre-existing should remain
.claude/skills/gaia-testing/SKILL.md, line 218git -C <repo> worktree add /tmp/ev -b <branch>-evidence origin/main
.claude/skills/gaia-testing/SKILL.md, line 228rclone copy <run-dir>/shots/NN_step.png gaia:amd-gaia/testing/<pr-or-issue>/<run-id>/ --s3-no-check-bucket

Connects to the internet

This is disclosed by the author in the skill's SKILL.md or README.

Found in 1 place.

.claude/skills/gaia-testing/SKILL.md, line 207curl -s -o /dev/null -w "%{http_code}\n" <pre-existing-service-url> # pre-existing still healthy

Downloads and installs other software

The author does not mention this.

Found in 3 places.

CLAUDE.md, line 430several worktrees that install points at whichever one ran `pip install -e` last.
CLAUDE.md, line 522uv venv && uv pip install -e ".[dev]"
CLAUDE.md, line 523uv pip install -e ".[ui]" # For Agent UI development

Safer alternatives

Low-risk skills in the same category (design) that look related. Check that one does what you need.

Install

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

if [ -e ~/.agents/skills/'gaia-testing' ] || [ -L ~/.agents/skills/'gaia-testing' ]; then echo 'A skill folder named gaia-testing 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 "$d" checkout --quiet '3ec6ceb6429b99c4af63250689239478a4b06c34' -- '.claude/skills/gaia-testing' && mkdir -p ~/.agents/skills && cp -R "$d"/'.claude/skills/gaia-testing' ~/.agents/skills/'gaia-testing' && 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/'gaia-testing' ~/.claude/skills/'gaia-testing' (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 badge for amd/gaia

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

Score history

Date	Version	Score
2026-11-13	`3ec6ceb`	79
2026-11-09	`2abaa51`	79
2026-11-07	`c11ea99`	79