persona-project-manager
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts on data from external sources such as Gmail, Google Drive, and Google Sheets while maintaining broad write capabilities, which creates a potential surface for indirect prompt injection.
- Ingestion points: The skill reads information via
gws workflow +weekly-digest(Gmail and Calendar) and searches for project artifacts usinggws drive files list. - Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between user instructions and potentially malicious instructions embedded in incoming emails or documents.
- Capability inventory: The skill possesses write capabilities including sending emails (
gws gmail +send), posting to chat (gws-chat), appending to spreadsheets (gws sheets +append), and modifying calendars (gws calendar +insert). - Sanitization: The instructions do not specify any validation or sanitization steps for the agent to perform on external data before using it to generate messages or updates.
Audit Metadata