# Promptfoo

> Test what your AI app actually answers — Test and compare what your prompts, agents and RAG apps actually answer.

- Page: https://tesign.com/en/item/promptfoo/
- JSON: https://tesign.com/en/item/promptfoo/index.json
- Korean Markdown: https://tesign.com/item/promptfoo/index.md
- Generated: 2026-09-21 05:13 UTC

## Numbers

- 25,177 stars — checked on GitHub 2026-09-16 13:42 UTC
- 7-day +5 observed via GH Archive (as of 2026-09-21 00:00 UTC)
- Star total = the value last checked on GitHub (stars_checked_at) + increases observed via GH Archive since. The 24h · 7d · 30d gains are GH Archive hourly events summed to the reference time (as_of). No score of ours.

## AT A GLANCE

- LICENSE: MIT (permissive) — https://spdx.org/licenses/MIT.html
- USAGE: Use, change and redistribute, commercially too. Keep the notice.
- OPEN SOURCE: YES
- LANGUAGE: TypeScript
- PLATFORM: cli
- CATEGORY: AI · DEV TOOLS
- Tags: ai 시험 · 프롬프트 · 레드팀 · 개발자용
- How to start: Install to use
- SOURCES: GitHub https://github.com/promptfoo/promptfoo
- INSTALL: https://promptfoo.dev/

## ACTIVITY

- Last commit: 2026-09-16 11:21 UTC
- Latest release: [unconfirmed]
- Contributors: [unconfirmed]
- Open issues (incl. PRs): [unconfirmed]
- Made by: [unconfirmed]
- Checked on GitHub: 2026-09-16 13:42 UTC

## TESIGN TAKE

It replaces "that feels better" with a table you can point at. The repository states plainly that it joined OpenAI and stays open source under MIT.

## WHY IT MATTERS

Every prompt change ends up judged by feel. Promptfoo runs the same inputs across models and prompt variants and lays the answers side by side in a matrix, with red-teaming for vulnerabilities in the same tool.

## BUILD FROM THIS

- A CLI and a library: install with npm, brew or pip, or run it with npx and install nothing. Provider keys go in as environment variables, results open as a browser matrix, and it drops into CI as a regression check.

## WHO IT'S FOR

- People shipping AI features, and teams that must keep quality steady while the prompts keep changing.

## START IN 5 MINUTES

```
# Create an example with npx, run the eval and open the results view.
```

## CAVEATS

- Needs Node.js 22.22 or newer. Calls run on your own provider keys, so a large eval costs real money there, and there is a paid enterprise edition alongside.

## RECEIPT

- FIRST SEEN BY TESIGN: 2026-09-16 13:41 UTC
- AT SOURCE: 2023-04-28 15:48 UTC
- KEPT: 2026-09-21 05:13 UTC
- Published on TESIGN: 2026-09-21 05:13 UTC
- Text last updated: 2026-09-21 05:13 UTC

## How to read this

- This file was produced by the same build, from the same data, as the tesign.com item page. The text is editorial; the numbers are stored observations.
- null in the JSON means not observed — never zero. The Markdown writes [unconfirmed] for it.
- Star total = the value last checked on GitHub (stars_checked_at) + increases observed via GH Archive since. The 24h · 7d · 30d gains are GH Archive hourly events summed to the reference time (as_of). No score of ours.
- A summary, not legal advice.

[Image] https://tesign.com/img/promptfoo-ae5c81e911.png