Evaluates skill routing accuracy for the suggest_skill keyword-matching logic. Tests both the MCP tool's keyword scoring and the UserPromptSubmit hook's grep-pattern detection, using a labeled dataset of (query → expected_skill) pairs.
Whether the UserPromptSubmit hook should fire for this query
Evaluates skill routing accuracy for the suggest_skill keyword-matching logic. Tests both the MCP tool's keyword scoring and the UserPromptSubmit hook's grep-pattern detection, using a labeled dataset of (query → expected_skill) pairs.