Write an eval suite for the pf-ai-experience-patterns skill (design-guide plugin). This skill applies Red Hat's AI design language to AI-powered features — chatbots, assistants, generation UIs.
Discriminating signal: Red Hat-specific AI UX patterns that baseline Claude wouldn't know — chatbot interaction models, generation UI patterns, assistant affordances per the RH AI design language. Baseline Claude may suggest generic AI UX patterns that don't align with RH brand.
MCP dependency: None — full eval with UI scenario descriptions.
Acceptance criteria:
-
eval/pf-ai-experience-patterns/eval.yaml exists following the pf-unit-test-generator template
-
5+ test cases covering explicit invocation, implicit/contextual prompts, and at least 1 negative control (e.g., generic form UI that shouldn't trigger AI patterns)
-
Inline Python judges test discriminating behavior (RH AI-specific recommendations, not general chatbot/AI UX advice)
-
With-skill/without-skill A/B delta documented
-
All judges pass at defined thresholds
Jira Issue: PF-4341
Write an eval suite for the
pf-ai-experience-patternsskill (design-guideplugin). This skill applies Red Hat's AI design language to AI-powered features — chatbots, assistants, generation UIs.Discriminating signal: Red Hat-specific AI UX patterns that baseline Claude wouldn't know — chatbot interaction models, generation UI patterns, assistant affordances per the RH AI design language. Baseline Claude may suggest generic AI UX patterns that don't align with RH brand.
MCP dependency: None — full eval with UI scenario descriptions.
Acceptance criteria:
eval/pf-ai-experience-patterns/eval.yamlexists following the pf-unit-test-generator template5+ test cases covering explicit invocation, implicit/contextual prompts, and at least 1 negative control (e.g., generic form UI that shouldn't trigger AI patterns)
Inline Python judges test discriminating behavior (RH AI-specific recommendations, not general chatbot/AI UX advice)
With-skill/without-skill A/B delta documented
All judges pass at defined thresholds
Jira Issue: PF-4341