muse

Generate a retrospective markdown document for archived mumei features from cost-log.jsonl.

2|1|Updated May 2, 2026
One-click install
npx skills add https://github.com/iroha924/mumei --skill muse-iroha924
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: muse
Source: https://github.com/iroha924/mumei/tree/main/skills/muse
Command: npx skills add https://github.com/iroha924/mumei --skill muse-iroha924

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires bash, CLAUDE_PLUGIN_ROOT, and includes scripts (resource) components.

What problem does it solve?

This Skill automates the generation of retrospective markdowns for archived mumei features, ensuring team knowledge retention and enhancing feature evaluation.

Core Features & Use Cases

  • Feature Retrospective: Generates markdown documents summarizing a finished mumei feature.
  • Reads multiple sources: Consumes data from requirements, design, tasks, spec-reviews, reviews, and cost-logs.
  • Use Case: After a feature is shelved, invoking /mumei:muse <feature> creates a comprehensive muse.md file that can be reviewed for future reference.

Quick Start

Generate a retrospective markdown for the feature 'feature-name' by running: /mumei:muse feature-name

Frequently Asked Questions about muse

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

FAQPage Schema
How do I generate a feature retrospective in markdown after archiving a project?

To generate a feature retrospective in markdown, run the `/mumei:muse <feature-name>` command. This creates a comprehensive `muse.md` file summarizing the archived feature's metrics, patterns, and lessons learned for future reference.

What sources are needed to automate a feature retrospective for team knowledge retention?

Automating a feature retrospective requires access to feature-related documents, including requirements, design, tasks, spec-reviews, reviews, and the `cost-log.jsonl` file. These sources provide the data needed to analyze and summarize the archived feature.

When should I generate a retrospective markdown for a completed feature?

You should generate a retrospective markdown immediately after a feature is shelved or archived. This ensures team knowledge retention by documenting metrics, patterns, and lessons learned while the feature context is still fresh.

Does the retrospective generation process require any specific environment dependencies?

Yes, generating the retrospective markdown requires a bash environment and access to the `CLAUDE_PLUGIN_ROOT`. These dependencies support the scripts that analyze feature documents and the cost-log.jsonl file to produce the output.

Can I use this retrospective tool for features without a cost-log file?

No, the `cost-log.jsonl` file is a required input for analysis. Without it, the tool cannot generate a comprehensive retrospective markdown containing the expected metrics, patterns, and lessons learned for the archived feature.

What is the best way to summarize an archived feature's metrics and lessons learned?

The best way to summarize metrics and lessons learned is to automate the process using a retrospective tool that reads multiple project documents. This generates a structured markdown file capturing patterns and cost data for future evaluation.