onboard
Pass
Audited by Gen Agent Trust Hub on Jul 19, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from connected tools such as email and Slack to build user profiles. Ingestion points: Phase 2 discovery scan reads content from communication tools (Email, Slack) and document tools (Drive, Notion). Boundary markers: The skill implements a 'Context Validation Gate' where findings are presented for user approval before being saved. Capability inventory: The skill can write files to the '~/Desktop/OS/' directory and create scheduled triggers via the '/schedule' skill. Sanitization: All generated context files and instructions are previewed and confirmed by the user before writing.
Audit Metadata