The Agent Skills Directory

[SAFE]: The skill is a text-based instructional guide. It focuses on pedagogical interaction and does not involve any technical operations that could pose a security risk. It does not use external tools, perform network operations, or execute code.
[PROMPT_INJECTION]: While the skill processes user-provided code, which represents an ingestion point for external data, the capability inventory is restricted to purely conversational output. No boundary markers or sanitization are explicitly defined, but the lack of dangerous capabilities (no file system writes, no network access, no tool usage) renders the injection surface safe.

refactoring-coach