recipe-forward-labeled-emails
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill is designed to forward email content to an external recipient address (
manager@company.com). While this is the primary purpose of the recipe, it involves the transfer of potentially sensitive internal communication to a third-party destination. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by reading email content through the
gws gmail users messages getcommand. This creates a vulnerability surface where instructions embedded in a received email could potentially influence the agent's behavior during the forwarding process. - Ingestion points: Email content retrieved via
gws gmail users messages getinSKILL.md. - Boundary markers: None identified; the skill does not use delimiters to wrap the fetched content.
- Capability inventory: The skill uses
gws gmail +forwardwhich performs a network-based write operation. - Sanitization: No sanitization or filtering of the email body is performed before the forward action.
Audit Metadata