bmad-retrospective

Run a post-epic retrospective that synthesizes story patterns into lessons learned and action items.

4|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/curdx/curdx-flow --skill bmad-retrospective-curdx
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bmad-retrospective
Source: https://github.com/curdx/curdx-flow/tree/main/.agents/skills/bmad-retrospective
Command: npx skills add https://github.com/curdx/curdx-flow --skill bmad-retrospective-curdx

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams transform the messy aftermath of an epic into a structured retrospective that captures lessons learned, evaluates success, and produces actionable next steps without blame.

Core Features & Use Cases

  • Epic discovery and verification: Detects the most relevant completed epic by checking sprint status first, then falls back to user input or story artifacts when needed.
  • Deep story synthesis: Reviews story records to extract recurring themes across dev notes, review feedback, takeaways, technical debt, and testing signals.
  • Continuity with prior retros: Loads the previous epic’s retrospective (when available) and cross-references action items, lessons, process changes, and debt follow-through.
  • Next-epic preparation: Previews the next epic to identify dependencies, gaps, and technical prerequisites so improvements connect directly to upcoming work.
  • Facilitated, psychologically safe dialogue: Runs a two-part format (Epic Review + Next Epic Preparation) with a “party mode” dialogue structure for authentic team interaction and user participation.

Quick Start

Use the bmad-retrospective Skill when your goal is to run a non-blaming post-epic review and produce lessons learned plus clearly owned action items, for example: “Let’s run a retrospective for the epic I just completed.”

Frequently Asked Questions about bmad-retrospective

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a post-epic retrospective that turns completed work into actionable improvements?

A post-epic retrospective synthesizes story records, dev notes, and review feedback to extract recurring themes and produce clearly owned action items. It evaluates epic completion success and generates system-focused lessons learned without assigning blame.

How do I synthesize technical debt and recurring patterns from story artifacts after an epic?

Synthesizing technical debt from story artifacts involves reviewing dev notes, testing signals, and takeaways to extract recurring themes. This process evaluates completion success and generates actionable next steps while maintaining continuity with prior retrospectives.

What is the best way to prepare for the next epic using lessons learned from the previous one?

Next-epic preparation previews upcoming work to identify dependencies, gaps, and technical prerequisites. It connects improvements directly to future tasks by cross-referencing prior action items, process changes, and debt follow-through.

Can I use previous retrospective action items to maintain continuity across multiple epic reviews?

Previous retrospective action items maintain continuity by loading prior lessons and cross-referencing debt follow-through. This verifies that past process changes were implemented and connects completed work directly to upcoming technical prerequisites.

Does a structured retrospective format help with psychologically safe team facilitation?

A structured retrospective format enforces a psychologically safe, system-focused dialogue using a two-part epic review and next-epic preparation structure. This approach encourages authentic team interaction and user participation while focusing on system outcomes.

When do I need to verify epic completion status before running a retrospective?

Epic completion verification is needed before running a retrospective to ensure the work is actually finished. The process checks sprint status first, then falls back to user input or story artifacts to detect the most relevant completed epic.