recipe-reschedule-meeting
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection vulnerability surface. Ingestion points: Untrusted data enters the agent context in SKILL.md when retrieving event details using the 'gws calendar events get' command, as event descriptions or titles can be modified by external organizers or attendees. Boundary markers: Absent; there are no clear delimiters or instructions telling the model to disregard embedded commands in the calendar fields. Capability inventory: The skill includes capabilities to modify entries and blast updates to all attendees via 'gws calendar events patch'. Sanitization: Absent; no validation or escaping of the fetched event details is implemented before processing.
Audit Metadata