Add an integration test that runs the full evaluate() function end-to-end.
In packages/core/src/__tests__/, create a test file that:
- Loads a fixture CV from
__fixtures__/
- Optionally loads a fixture JD
- Calls
evaluate() with both
- Asserts the result has all required fields (score, dimensions, issues, strengths, archetype)
- Asserts the score is within a reasonable range
- Asserts known issues are detected (e.g., a CV with 'passionate about AI' should flag it)
- Asserts archetype detection works (a Python/ML heavy CV should detect as ai-engineer)
Use vitest. This is the foundation for the golden fixture test harness described in ARCHITECTURE.md.
Add an integration test that runs the full
evaluate()function end-to-end.In
packages/core/src/__tests__/, create a test file that:__fixtures__/evaluate()with bothUse vitest. This is the foundation for the golden fixture test harness described in ARCHITECTURE.md.