The Agent Skills Directory

[SAFE]: The skill consists of purely instructional content aimed at improving the reliability and safety of code refactoring tasks performed by AI agents. It includes detailed checklists for various scenarios such as table refactoring, state management extraction, and circular dependency prevention.
[SAFE]: No evidence of prompt injection, data exfiltration, or malicious command execution was found. The instructions encourage the agent to use standard tools like Read and Grep to gather necessary context before making changes, which is a best practice for coding assistants.
[SAFE]: The examples provided for Java, Go, and TypeScript are illustrative and focus on identifying and fixing engineering anti-patterns such as 'Shotgun Fixes' and positional list de-synchronization.

cc-refactor-safety