The Agent Skills Directory

[SAFE]: The skill is a purely instructional framework for privacy auditing and does not include any executable code or scripts.
[SAFE]: No remote code execution, network operations, or sensitive file access patterns were identified. All grep patterns provided are for identification of privacy leaks in analyzed code, not for data exfiltration.
[SAFE]: The skill specifies disable-model-invocation: true, which is a security restriction that prevents the agent from calling other models, limiting the potential attack surface.
[SAFE]: The skill references local relative paths (e.g., ../../shared/schemas/flags.md) for configuration and framework documentation, which is standard for modular skill repositories.

data-disclosure