Promptfoo is an open-source LLM security platform designed to help developers secure AI applications from prompt to production. It offers features for red-teaming, guardrails, model security, and evaluations. Key features include adaptive red teaming, customized attack generation, and specialized language models for identifying vulnerabilities. It supports various deployment options, including CLI tools, managed cloud, and on-premises enterprise solutions. Use cases include identifying prompt injections, jailbreaks, data leaks, insecure tool use, and toxic content generation. It is used by Fortune 500 companies and has a large open-source community.
Promptfoo
Open-source LLM security platform for red-teaming, guardrails, model security, and evaluations, trusted by 150,000+ developers.
