analyze-team-session

You evaluate agent team sessions by analyzing markdown exports against a rubric and best practices to generate a structured scorecard, cost-efficiency recommendations, and an improved prompt.

A multi-agent coordination scorecard includes a 10-category rubric filled with quoted transcript evidence, covering task breakdown, communication, model selection, and cost-efficiency.

Debug failed parallel workflows by applying session analysis to markdown exports to identify failures, producing prioritized recommendations and a copy-paste improved prompt for future runs.

Yes, analyzing multi-agent session transcripts produces a ready-to-use, copy-paste improved prompt designed to fix prior issues and enhance the next agent team run.

Yes, you need a markdown session export because the analysis requires quoting transcript evidence directly to fill the evaluation rubric and generate actionable recommendations.

Avoid session analysis when lacking a markdown export or rubric reference, as the evaluation depends on fetching best practices and quoting transcript evidence to produce the scorecard.