promptfoo-evals

80 / 100 · Low risk

Fine to try. Some of what it can do is expected: a devops skill usually runs commands, installs tools and uses cloud credentials.

By promptfoo/promptfoo (GitHub) · devops · 26,951 stars · scanned at commit 2a6fdf9

Write, run, and improve non-redteam Promptfoo eval suites for a configured target: test cases, assertions, rubrics, datasets, and CI gates. Use promptfoo-provider-setup first for a new or broken connection; use the redteam skills for adversarial scans.

Works with any agent that supports SKILL.md skills.

What this skill can do

Downloads and installs other software

The author does not mention this. Expected for a devops skill.

Found in 9 places.

plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 62Use `npx promptfoo` to resolve the project's installed CLI and record its version. Install or upgrade
plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 63with `npx promptfoo@latest` only when needed. In the Promptfoo repository, align
plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 65of `npx promptfoo` below.

Show 6 more

plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 68npx promptfoo validate config -c path/to/promptfooconfig.yaml
plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 69npx promptfoo eval -c path/to/promptfooconfig.yaml -o output.json --no-cache --no-share
plugins/promptfoo/skills/promptfoo-evals/references/eval-patterns.md, line 224npx promptfoo eval -c promptfooconfig.yaml --filter-pattern invoice -o /tmp/invoice.json --no-cache --no-share
plugins/promptfoo/skills/promptfoo-evals/references/eval-patterns.md, line 225npx promptfoo eval -c promptfooconfig.yaml --filter-metadata area=billing -o /tmp/billing.json --no-cache --no-share
plugins/promptfoo/skills/promptfoo-evals/references/eval-patterns.md, line 226npx promptfoo eval -c promptfooconfig.yaml --filter-failing /tmp/eval-results.json -o /tmp/failing.json --no-cache --no-share
plugins/promptfoo/skills/promptfoo-evals/references/eval-patterns.md, line 237PROMPTFOO_FAILED_TEST_EXIT_CODE=0 npx promptfoo eval -c promptfooconfig.yaml -o "$result_dir/results.json" --no-cache --no-share

Runs commands on your computer

The author does not mention this. Expected for a devops skill.

Found in 1 place.

plugins/promptfoo/skills/promptfoo-evals/SKILL.md, line 68npx promptfoo validate config -c path/to/promptfooconfig.yaml

Other copies in this repository

promptfoo/promptfoo has 1 more folder with a skill named promptfoo-evals. This page shows the lowest-scoring copy.

.claude/skills/promptfoo-evals 82/100 · Low risk

Install

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

if [ -e ~/.agents/skills/'promptfoo-evals' ] || [ -L ~/.agents/skills/'promptfoo-evals' ]; then echo 'A skill folder named promptfoo-evals already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/promptfoo/promptfoo' "$d" && git -C "$d" checkout --quiet '2a6fdf9b2e1600ffa2a204e70e9fea7cb0c14cff' -- 'plugins/promptfoo/skills/promptfoo-evals' && mkdir -p ~/.agents/skills && cp -R "$d"/'plugins/promptfoo/skills/promptfoo-evals' ~/.agents/skills/'promptfoo-evals' && 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/'promptfoo-evals' ~/.claude/skills/'promptfoo-evals' (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 promptfoo/promptfoo shows the score of its lowest-scoring skill.

SkillTrust badge for promptfoo/promptfoo

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

Score history

Date	Version	Score
2026-11-08	`2a6fdf9`	80