The Agent Skills Directory

[SAFE]: The skill serves as an instructional template for refactoring tasks, guiding the agent through codebase analysis and planning. It does not introduce malicious scripts, external dependencies, or hidden commands.
[INDIRECT_PROMPT_INJECTION]: The skill uses the {{refactor_description}} variable in SKILL.md to ingest untrusted data into the prompt. This establishes a surface for indirect prompt injection where malicious instructions could be embedded in the refactor goal. The skill lacks explicit boundary markers or sanitization for this input; however, the risk is minimal as the output is a plan intended for human review and the capabilities involved (codebase searching) are consistent with the skill's purpose.

refactor-plan