Spaces:
Sleeping
Sleeping
Commit History
-
Add PC Pal evaluation framework with structural metrics and rubric scoring
Implements eval/ directory with 10 automated structural metrics (no API needed), 6 rubric dimensions with precomputed scores for 6 sample conversations, a CLI runner supporting precomputed and interactive modes, and result/summary JSON output. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>980f6ba
