The Agent Skills Directory

[SAFE]: The skill instructions define a legitimate and secure application security review process using source-to-sink analysis. No prompt injection or behavior override patterns were detected.
[SAFE]: The skill's allowed tools are limited to read-only operations (Read, Grep, Glob), which is appropriate for its intended purpose and minimizes security risks.
[SAFE]: Language-specific references and examples are used to educate the agent on vulnerability patterns and do not involve remote code execution, unauthorized network calls, or exfiltration of sensitive data.

security-review