The Agent Skills Directory

[SAFE]: The skill instructions are entirely focused on improving code structure through refactoring. No malicious intent, obfuscation, or unauthorized behaviors were identified during the analysis.
[COMMAND_EXECUTION]: The skill utilizes basic shell commands (e.g., mkdir, touch) and suggests running standard project verification commands (e.g., npm run typecheck, tsc). These actions are restricted to the local project environment and are consistent with the skill's primary purpose of code reorganization.
[PROMPT_INJECTION]: As the skill involves reading and processing external source code files, it possesses an attack surface for indirect prompt injection (Category 8). However, the risk is assessed as safe because the skill provides prescriptive, step-by-step instructions that limit the agent's autonomy and do not involve exfiltrating data to external sources.

file-refactor