End-to-end tests for your Python AI app.
What it does
Builds an end-to-end evaluation pipeline for Python LLM applications that produces real scores and an improvement plan.
When to use it
Use when asked to add tests or evaluations, benchmark, fix incorrect behavior, or improve quality in a Python project that calls an LLM.
How to use it
Give it a Python LLM application to evaluate; it builds and runs an evaluation pipeline and saves scored results, analysis, and an action plan.
What you provide
- An existing project
Good to know
- cannot be undone: eval-driven-dev
- acts without asking
- edits its own instructions
Latest review
What you need · 1
Requires Python 3.10 or newer.
Access · 4
This skill
eval-driven-dev
Write
Updates its own skill instructions (irreversible)
npx skills
Execute
The skills updater
pixie-qa
Execute
Installs or upgrades pixie-qa