retrospective

Generates sprint or milestone retrospectives by analyzing completion data, velocity, and blockers.

13|Updated Jul 30, 2026
One-click install
npx skills add https://github.com/frabcd/codex-ai-game-studio --skill retrospective-frabcd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: retrospective
Source: https://github.com/frabcd/codex-ai-game-studio/tree/main/plugins/ai-game-studio/skills/retrospective
Command: npx skills add https://github.com/frabcd/codex-ai-game-studio --skill retrospective-frabcd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often skip retrospectives or write vague ones because gathering sprint data, computing velocity, and tracking recurring blockers is tedious manual work. This Skill automates the analysis and produces a structured, data-backed retrospective document. ## Core Features & Use Cases - Automated Data Collection: Reads sprint plans, milestone definitions, sprint-status YAML, and git history to compare planned versus actual work. - Trend and Debt Analysis: Tracks velocity across sprints, counts TODO/FIXME/HACK comments, and follows up on previous action items. - Structured Output: Produces a markdown retrospective with metrics tables, blocker analysis, estimation accuracy, and prioritized action items saved to production/retrospectives/. - Use Case: After finishing sprint 4 of a game project, run the Skill to generate a retrospective showing a 75% completion rate, a growing TODO count, and three carryover tasks, then feed the action items directly into the next sprint plan. ## Quick Start Use the retrospective skill to generate a retrospective for sprint 4 and save it to production/retrospectives/.

Frequently Asked Questions about retrospective

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

FAQPage Schema
How do I generate a sprint retrospective automatically?

Run the retrospective skill with a sprint identifier like sprint-4. It reads the sprint plan, sprint-status.yaml, and git log for the period, then produces a markdown retrospective with metrics, blockers, and action items saved to production/retrospectives/.

How to track velocity trends across multiple sprints?

The skill reads previous retrospectives from production/retrospectives/ and builds a velocity trend table comparing planned versus completed tasks across recent sprints. It labels the trend as increasing, stable, or decreasing with a one-sentence explanation.

What happens if no sprint data exists for the retrospective?

The skill reports that no sprint data was found and offers two options: provide the sprint tasks, dates, and outcomes manually, or stop with a BLOCKED verdict. Manual data is then used as the source of truth for the analysis.

Can I update an existing retrospective instead of creating a new one?

Yes. The skill globs for existing retrospective files matching the sprint or milestone name and prompts you to either update the existing file with new data or archive it with a date suffix and start fresh.

Does the retrospective skill track technical debt?

Yes. It counts TODO, FIXME, and HACK comments in the codebase, compares them against counts from previous retrospectives, and reports whether technical debt is growing, stable, or shrinking.