meeting-scheduler
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, creating a potential surface for indirect prompt injection. \n
- Ingestion points: The agent ingests natural language requests from users and calendar data (titles, descriptions, attendee lists) from external APIs such as Google Calendar, Microsoft Graph, and Calendly. \n
- Boundary markers: Absent; there are no specific instructions or delimiters provided to help the agent distinguish between untrusted calendar content and core system directives. \n
- Capability inventory: The agent has the capability to write to calendar APIs and send emails with .ics attachments based on processed inputs. \n
- Sanitization: Absent; no validation or escaping mechanisms are defined for the data retrieved from external sources or user inputs. \n- [NO_CODE]: The skill consists entirely of markdown documentation and instructional content. It does not include any scripts, binaries, or configuration files that would be executed on the host system.
Audit Metadata