Powerbi AI Frown To Fix Triage

Power BI AI Frown-to-Fix Triage

Use this skill to translate Power BI Desktop Frown output into a concrete repair plan.

Triage Pattern

1. Capture error message, timestamp, Desktop version, session id, and failing artifact.
2. Classify failure domain: PBIP/PBIR structure, TMDL model, relationship graph, DAX, Power Query/M, native source, DirectQuery/live connection, visual binding, or Desktop compatibility.
3. Extract named tables, measures, relationships, queries, pages, and visuals.
4. Map symptoms to likely source files and validators.
5. Produce a fix plan and retest command sequence.

Output Requirements

• Include failure class, evidence, likely root cause, affected files, fix plan, and retest path.
• Do not guess silently; mark low-confidence diagnoses and required evidence.

Executable Support

• Use scripts/powerbi_desktop_smoke_test.ps1 to capture before/after Frown evidence.
• Use scripts/powerbi_expert_factory.py validate first to catch common causes before reopening Desktop: ambiguous paths, missing visual bindings, brittle M navigation, and unsupported demo source patterns.