The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard command-line tools including git, gh (GitHub CLI), grep, find, and wc to perform code analysis and history tracking. These tools are used for their intended purpose in a developer workflow.
[PROMPT_INJECTION]: No malicious injection patterns or safety bypass attempts were found. The skill provides structured guidance for the agent's behavior as a security auditor without overriding system instructions.
[DATA_EXFILTRATION]: No unauthorized data transfer mechanisms or credential harvesting patterns were detected. The skill focuses on analyzing local code and generating reports locally.
[EXTERNAL_DOWNLOADS]: The skill does not download external scripts or execute remote code. It references other local agent skills (audit-context-building, issue-writer) for integrated workflows.
[SAFE]: The skill implements a robust, multi-phase security auditing methodology with clear evidence-based requirements. All identified patterns are consistent with the primary purpose of performing security reviews.

differential-review