Back to radar

Production AI Radar

Promptfoo

CLI/CI evals for prompts, RAG, red-team, and provider comparison.

TrialLLMOpsNew
Why this ring
Fastest path to eval gates in CI; complements Langfuse for trace-level debug.
Production risk if ignored
Eval suite too small - passes CI but fails in production edge cases.
Typical effort
days
Low FinOps impact

Use cases

  • Prompt regression CI
  • Provider bake-offs
  • Red-team scenarios

Adoption steps

  1. Add promptfoo.yml to repo
  2. Block merge on regression
  3. Expand golden set monthly

Related tools

In your assessment

CI eval gate coverage review