Meeting observation to email reminder
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing external data during its operation.
- Ingestion points: Meeting details are retrieved from QueryMeeting observations, and recipient email addresses are extracted from user dialogue (SKILL.md, rules.md).
- Boundary markers: There are no explicit delimiters or boundary markers specified to isolate meeting data from the instructions, though the use of a fixed content template reduces the risk of unintended instruction execution.
- Capability inventory: The skill invokes the EmailReminder tool to process and send the final content (validator.py).
- Sanitization: No sanitization or validation of meeting_name, meeting_location, or meeting_time is performed before these values are interpolated into the deterministic email content template.
Audit Metadata