FICHA · AUR

promptfoo

Test and evaluate LLM outputs - AI red teaming, pentesting, and vulnerability scanning

  • LLM security evaluation tool
  • CLI
  • NETWORK
  • Launchable
  • Runs in terminal
official+codex · reviewed · Jun 3, 2026 description in en

Description

LLM applications can be tested for output quality, jailbreaks, prompt-injection behavior, and other security weaknesses. This developer security tool evaluates prompts and model responses for AI red teaming, pentesting, and vulnerability scanning. Test cases may send prompts, data, and outputs to model providers, so authorization, privacy, and target scope are critical.

How to run

promptfoo

Commands: promptfoo

Permissions

Permissions not analysed for this source yet.