Promptfoo is a security testing platform for AI applications that brings vulnerability checks into your development workflow. It helps teams evaluate agents, chatbots, and other LLM-based apps for attack resilience before release.
AI security in CI/CD
Promptfoo fits into existing engineering pipelines so tests can run automatically as code or prompts change.
- Integrates with GitHub, GitLab, Jenkins, and other CI/CD tools
- Supports MCP and multiple AI agent frameworks
- Deployable in the cloud or on-premise
Attack simulation and vulnerability discovery
The platform generates large volumes of context-aware attacks tailored to your specific application, using a real-world threat stream from a community of 200,000+ developers.
- Detects prompt injection, data leakage, and other common LLM risks
- Automates testing beyond what’s practical with manual scenarios
For product and enterprise teams
Promptfoo is used by companies including dozens in the Fortune 500, and is suitable for both startups and large teams building security- and compliance-sensitive AI services.
