Check your AI code with verified test results.
What it does
Sets up and runs offline evaluations of AI code, then verifies and reports their results in Logfire.
When to use it
Use for evaluation setup, datasets, AI scoring, agent checks, LLM judges, Braintrust migration, or Logfire experiments, not live monitoring.
How to use it
Give it an AI task in a project to evaluate; it runs evaluations and returns a report of verified results.
What you provide
- An existing project
Uses
Good to know
- cannot be undone: Logfire
What you need · 2
Requires authenticated access to the exact Logfire project for uploads, hosted datasets, or span evaluators; isolated local-only evaluations can omit authentication.
Routes authentication and exact-project selection for Logfire-backed native runs to this skill's linked reference; requiring installation of the whole skill is unstated.
Access · 2
This skill
Logfire
Uploads evaluations to Logfire (irreversible)
https://pydantic.dev/.well-known/agent-skills/logfire-instrumentation/references/auth.md
Reads Logfire authentication instructions