القائمة
Add or debug pytest unit, integration, and LLM evaluation tests for genai-tk, including fake models, real-model gates, and trajectory tests.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | evaluation-testing |
| description | Add or debug pytest unit, integration, and LLM evaluation tests for genai-tk, including fake models, real-model gates, and trajectory tests. |
docs/TESTING_GUIDE.mddocs/evaluation_testing.mdtests/conftest.pytests/utils/pyproject.toml dependency groups dev and evals| Tier | Path | Use for |
|---|---|---|
| Unit | tests/unit_tests/ | Fast deterministic behavior and config parsing |
| Integration | tests/integration_tests/ | CLI wiring, real services, browser, end-to-end flows |
| Eval | tests/eval_tests/ | LLM quality, trajectory, multiturn behavior |
| Utilities | tests/utils/ | Shared fake data, factories, constants |
GENAITK_PROFILE=pytest for deterministic tests.pytest.importorskip() for optional packages.uv run pytest tests/unit_tests -q
uv run pytest tests/integration_tests -q
uv run pytest tests/eval_tests -q
just test-unit
just test
Build or modify LangChain, DeepAgent, DeerFlow profiles, agent tools, middleware, checkpointing, skills wiring, and the shared harness layer in genai-tk.
Work on BAML structured extraction, BAML CLI commands, processors, utilities, and Prefect BAML workflow integration in genai-tk.
Work on browser automation, sandbox browser tools, direct Playwright tools, AioSandbox backend, and sandbox CLI support in genai-tk.
Add or modify genai-tk Typer CLI commands, dynamic command registration, project scaffolding, and generated Copilot/agent support files.
Work on genai-tk OmegaConf configuration, profiles, overrides, env substitution, and config discovery. Use when editing config/*.yaml or genai_tk.config_mgmt.config_mngr.
Work on core LLM, embeddings, vector store, provider, cache, prompt, and retriever factories in genai-tk. Use when editing genai_tk/core or provider configuration.