weekly-meeting-reminder
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill's operations are consistent with its stated goal of helping users set up a meeting summary automation. No malicious intent, obfuscation, or unauthorized data access patterns were detected.
- [COMMAND_EXECUTION]: The skill utilizes platform-provided tools such as
read_fileto process its instructions andcreate_scheduled_taskto establish the weekly automation. These tool calls are appropriate for the documented functionality. - [SAFE]: An indirect prompt injection surface is present because the skill processes external calendar data, but it is integral to the primary purpose of the skill. Evidence:
- Ingestion points: Google Calendar events via the
google_calendarintegration. - Boundary markers: Not present in the example instruction used for task creation.
- Capability inventory: File reading and scheduled task creation with email delivery capabilities.
- Sanitization: No specific sanitization or filtering of calendar content is described before inclusion in the summary email.
Audit Metadata