Open SourceLibraryFree

promptfoo

Declarative evals and red-teaming for prompts, agents, and RAG that run in CI, so a behavior regression fails a check instead of reaching users. Red-team module probes prompt injection, jailbreaks, PII leaks, and tool misuse. MIT. Note: OpenAI announced it is acquiring promptfoo (March 2026), so weigh the long-term open-source trajectory.

Homepage ↗Repo ↗

Alternatives

DSPy

1 workflow use promptfoo

AgentsFreeMachine-verified

promptfoo: make agent evals fail the build, not the user

Write a declarative promptfoo config with real assertions and wire promptfoo eval into CI, so a regression in prompt or agent behavior fails a check instead of reaching production.

promptfoo· intermediate