Modernize Legacy Systems with Confidence

Refactoring legacy code is a high-stakes endeavor where missing a single hidden dependency can lead to catastrophic failure. This skill transforms your AI agent into a methodical refactoring specialist, leveraging a battle-tested framework designed to modernize aging codebases while strictly preserving critical business logic.

What it does

This skill guides your AI through a multi-phase modernization strategy: mapping existing behaviors, generating characterization tests to capture current logic, and implementing the Strangler Fig pattern for incremental replacement. It ensures that every change is backed by a safety net before a single line of old code is deleted.

Supported Workflows

• Characterization Testing: Automatically capturing "ugly" but necessary behaviors as a baseline.
• Incremental Replacement: Moving from big-bang rewrites to safe, side-by-side migrations.
• Logic Extraction: Untangling complex business rules from spaghetti code.
• Risk Mitigation: Implementing strict safety rules for shared state and global variables.

Why use this skill?

Standard AI prompts often result in "hallucinated" improvements that break undocumented edge cases. This skill enforces a structured, safety-first methodology that treats legacy code with respect, ensuring your AI acts as a careful engineer rather than a reckless rewriter.