The Agent Skills Directory

[SAFE]: The skill operates exclusively on local project files and diffs provided by the user. It does not attempt to access sensitive system files, environment variables, or external network resources.
[SAFE]: No evidence of obfuscation, credential harvesting, or remote code execution was found. The instructions utilize built-in agent capabilities for task delegation and architectural planning.
[SAFE]: The workflow includes explicit auditing steps (Step 2) to verify findings before application, acting as a control against unintended code modifications or indirect prompt injection from comments within the processed files.

no-comments