recipe-cancel-and-notify
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from an external source (Google Calendar events) and uses that information to perform subsequent actions like deleting events and sending emails. This flow represents a vulnerability surface where malicious content embedded in a calendar event could attempt to influence the agent's behavior.
- Ingestion points:
gws calendar +agendaoutput (SKILL.md) - Boundary markers: None present to delimit or ignore instructions within external data.
- Capability inventory:
gws calendar events delete,gws gmail +send(SKILL.md) - Sanitization: No explicit sanitization or validation of the event data is described before use in email bodies or command parameters.
- [COMMAND_EXECUTION]: The skill instructions include the execution of the
gwsbinary to interact with Google Workspace services. This usage is consistent with the skill's stated purpose of managing calendar events and sending emails.
Audit Metadata