Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a comprehensive security model designed to mitigate risks from untrusted content. It explicitly treats all message content, attachments, and external web content as untrusted, requiring independent, trusted authorization for any 'send' operation.
- [SAFE]: Data integrity is maintained through the use of 'protected facts' and SHA-256 artifact hashing. The skill ensures that critical information like names, addresses, and URLs remains unchanged throughout the drafting and rendering process.
- [SAFE]: The skill uses robust path traversal protections when accessing files within message bundles. It employs path resolution and relative-to-root checks to ensure all artifact access is confined to authorized directories.
- [SAFE]: HTML generation is performed deterministically using a dedicated script that applies strict escaping to all user-provided text. Style attributes are constrained to a predefined, validated set of policy tokens, preventing CSS injection or malicious styling.
- [SAFE]: The skill adheres to the principle of least privilege and transport neutrality. It does not implement network protocols itself and explicitly forbids the storage of credentials within its environment or message bundles.
Audit Metadata