gds-retrospective

Facilitates post-epic retrospectives to extract lessons learned and prepare for the next epic.

1|Updated Sep 18, 2026
One-click install
npx skills add https://github.com/PastaSus/egg-defender --skill gds-retrospective-pastasus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gds-retrospective
Source: https://github.com/PastaSus/egg-defender/tree/main/.agents/skills/gds-retrospective
Command: npx skills add https://github.com/PastaSus/egg-defender --skill gds-retrospective-pastasus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After completing an epic, teams often skip structured reflection, losing valuable lessons from story implementation, code reviews, and technical debt. This Skill runs a guided, multi-agent retrospective that analyzes story records, checks follow-through on previous action items, and produces a documented retrospective with actionable improvements. ## Core Features & Use Cases - Automated Epic Discovery: Detects the completed epic from sprint-status.yaml with fallback to story file scanning and user confirmation. - Deep Story Analysis: Mines dev notes, review feedback, lessons learned, and technical debt sections across all stories to surface recurring patterns. - Continuity Tracking: Cross-references the previous epic's retrospective to verify whether committed action items were completed. - Next Epic Preparation: Previews the upcoming epic, identifies dependencies on completed work, and flags preparation gaps. - Use Case: After finishing Epic 3 of a game project, run the retrospective to generate a blameless lessons-learned document, verify Epic 2 action items were applied, and produce a readiness checklist for Epic 4. ## Quick Start Run a retrospective for the epic we just completed and generate the lessons-learned document.

Frequently Asked Questions about gds-retrospective

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

FAQPage Schema
How do I run a retrospective for a completed epic?

Invoke the skill by saying "run a retrospective" or "lets retro the epic". It detects the completed epic from sprint-status.yaml, confirms with you, then guides a structured team discussion and produces a retrospective document.

How does the retrospective detect which epic was completed?

It reads sprint-status.yaml first to find the highest epic with stories marked done. If detection fails, it asks you directly, then falls back to scanning story filenames in the implementation artifacts folder.

Can I run a retrospective if some stories are not done?

Yes, but the skill warns that the epic is incomplete and lists pending stories. You can choose to finish remaining stories first or proceed with a partial retrospective, which is flagged as potentially missing lessons.

Does the retrospective check previous action items?

Yes. It loads the previous epic's retrospective and cross-references each committed action item against the current epic's story records, marking them completed, in progress, or not addressed.

What inputs does the retrospective workflow require?

It requires the agent manifest CSV and config.yaml for user settings, plus planning artifacts like epics, architecture, and GDD documents. Story files and sprint-status.yaml from implementation artifacts drive the analysis.

What are the limitations of this retrospective workflow?

It depends on well-maintained sprint-status.yaml and story files with dev notes sections. Without a previous retrospective or a defined next epic, continuity tracking and preparation steps are skipped.