Reche Inc
Reche Inc
ENChange languagePortuguêsBook a diagnosisChat on WhatsApp
Back to blog
AI Tools

Promptfoo: test, evaluate and red-team LLM apps

Promptfoo is an open-source tool to test and evaluate LLM applications, compare models and run security red-teaming, with simple configs and CI/CD integration.

Published on June 16, 20265 min readView on GitHub

Shipping AI to production without testing is a recipe for a bad surprise. Promptfoo solves that: it is an open-source tool to test, evaluate and red-team LLM apps, agents and RAG, with declarative config and direct CI/CD integration.

What is Promptfoo?

With simple YAML configs, Promptfoo compares models side by side (GPT, Claude, Gemini, DeepSeek), catches regressions before deploy and runs AI-specific vulnerability scanning and pentesting. It is used by hundreds of thousands of developers.

Key features

  • Side-by-side model comparison and regression detection
  • Red-teaming and vulnerability scanning for AI apps
  • Declarative YAML configs, MIT, with a native GitHub Action
  • First-class CI/CD integration
  • Around 22k stars and used by large companies

How Reche uses it

Automated evaluation is one of RecheOS's quality gates: no AI feature reaches production without testing and a security check. Reche implements that rigor in your product so AI is reliable, not a gamble.

Want to implement this in your product?

Reche's initial diagnosis defines scope, timeline, and budget. Credited to the project if you move forward.