bmad-retrospective

To run a structured retrospective, the skill discovers completed epics from sprint status, analyzes story evidence for lessons, and guides a no-blame team discussion. It requires YAML-frontmatter metadata and project configuration resolution to generate actionable next steps for the next epic.

Evidence-based analysis works by scanning story records to synthesize recurring struggles, review feedback patterns, breakthroughs, and technical debt. It extracts quality and test insights from story artifacts to ensure continuity from the previous epic and preparation for the next epic.

Yes, you need Python3 installed as a dependency to run the scripts that facilitate team discussions and extract lessons. The skill uses deterministic discovery logic and safeguarded facilitation outputs to analyze sprint status and story artifacts.

The best way to prepare continuity is by loading the previous epic's retrospective when available, cross-referencing commitments and lessons, and previewing the next epic's dependencies and risks. This ensures forward-looking planning is grounded in evidence from completed story artifacts.

You cannot use this skill effectively if stories are not marked done, because epic discovery and readiness checks verify whether stories are actually completed in the sprint status. The skill relies on completed story artifacts to synthesize recurring struggles and technical debt accurately.

Epic discovery and readiness check is the process of detecting the most relevant completed epic primarily from sprint status and verifying whether stories are actually marked done. It applies deterministic discovery logic to project configuration resolution and YAML-frontmatter metadata to ensure valid retrospective facilitation.