Streamlines test execution and strategy for the AGILAB AI engineering project using optimized pytest workflows.
The AGILAB Testing skill provides a specialized framework for validating changes within the AGILAB ecosystem, an AI-driven engineering initiative by Thales Group. It implements a 'start small' philosophy, prioritizing targeted unit tests and smoke tests for core modules to ensure rapid developer feedback without the overhead of full-scale regressions. By leveraging the uv package manager and pytest, this skill helps engineers maintain high code quality in AI experiments while avoiding common pitfalls like unrelated test failures or heavy resource dependencies.
Key Features
01Lightweight testing strategy avoiding GPU and large dataset dependencies
02Targeted unit testing for agi-env and core project modules
03Flexible command support for both local changes and full repository regressions
04Optimized execution using uv with preview features and no-sync flags
053 GitHub stars
06Quick smoke testing for application stability validation
Use Cases
01Adding narrow unit tests for path resolution and parsing helpers
02Validating local changes to AGILAB core logic before committing code
03Ensuring the AI engineering environment is correctly configured via smoke tests
