The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions designed to guide an AI agent through a security audit process. It does not include any executable scripts or binaries.
[SAFE]: No network exfiltration or sensitive data access patterns were identified. The workflow focuses on analyzing code provided by the user in a pull request or diff context.
[SAFE]: The instructions promote security best practices, such as redacting secret values and looking for vulnerabilities like prompt injection and insecure supply chains.

security-review