The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes local source code using discovery tools like Read, Grep, and Glob. This introduces a theoretical attack surface for indirect prompt injection if the project code being refactored contains hidden malicious instructions. However, the skill's methodology includes mandatory analysis steps and the execution of existing test suites, which serves as a natural verification process.
[SAFE]: The skill promotes the use of well-known, established security tools for Rails development, including strong_migrations for catching dangerous database operations and anchor_migrations for DDL lock protection. These recommendations align with professional software engineering safety standards.

rails-refactorer