migrate-decompose

Decompose migration manifests into granular features with verification criteria.

Updated Jan 10, 2026
One-click install
npx skills add https://github.com/dain-agency/mood-tracker --skill migrate-decompose
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: migrate-decompose
Source: https://github.com/dain-agency/mood-tracker/tree/main/.claude/skills/migrate-decompose
Command: npx skills add https://github.com/dain-agency/mood-tracker --skill migrate-decompose

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When migrating legacy applications, large source files are often tracked as single work items, leading to missing functionality even if the code compiles, because the work wasn't broken down into granular, trackable features with clear verification steps.

Core Features & Use Cases

  • Granular Feature Decomposition: Breaks down source application files into individual features sized for 30 minutes to 4 hours of work, with explicit source file line references and dependency mapping.
  • Auto-Detection of Completed Work: Automatically runs verification criteria against existing target files to mark already migrated features as complete, avoiding redundant work.
  • Prioritized Work Queue Generation: Produces a sorted list of remaining tasks ordered by priority and dependencies, so migration teams always know what to work on next.
  • Use Case: A team migrating a legacy Slack mood tracker bot can use this skill to break down components like Slash commands, Supabase integration, and daily check-in triggers into individual trackable tasks, ensuring no functionality is missed during the migration.

Quick Start

Use the migrate-decompose skill to break down your migration manifest for the mood-tracker Slack bot into individual, trackable feature tasks with clear verification criteria.

Frequently Asked Questions about migrate-decompose

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

FAQPage Schema
How do I break down a legacy application migration into trackable features?

To break down a legacy application migration, you can decompose source files into granular features sized for 30 minutes to 4 hours of work. This process assigns explicit source file line references and dependencies to prevent missing functionality during the migration.

How does feature decomposition handle already migrated source code components?

Feature decomposition handles already migrated components by automatically running machine-checkable verification criteria against existing target files. This auto-detection mechanism marks matching features as complete to prevent redundant work by the migration team.

What is the best way to generate a prioritized work queue for a migration project?

The best way to generate a prioritized work queue is to decompose your migration manifest into individual tasks with explicit dependencies. This produces a sorted list of remaining features ordered by priority, ensuring teams always know what to work on next.

Can I track legacy migration tasks without explicit source file line references?

Tracking legacy migration tasks without explicit source file line references is not recommended. The decomposition process enforces feature size constraints and requires explicit source file line ranges to ensure granular tracking and machine-checkable verification.

Why does legacy code compile but still have missing functionality after migration?

Legacy code compiles but has missing functionality because large source files are often tracked as single work items. Without decomposing the migration manifest into granular features with clear verification steps, individual tasks are easily overlooked.

Does migration decomposition work for migrating a Slack bot with Supabase integration?

Yes, migration decomposition works for a Slack bot with Supabase integration. It breaks down components like Slash commands, Supabase integration, and daily check-in triggers into individual trackable tasks, ensuring no functionality is missed.