recipe-review-meet-participants
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes conference and participant metadata retrieved from external Google Meet logs via the
gwscommand-line tool. - Ingestion points: Conference records, participant lists, and session details are retrieved in
SKILL.mdvia tool calls. - Boundary markers: Absent. The instructions do not define delimiters or specific safety instructions for the agent when processing the retrieved data.
- Capability inventory: The skill utilizes the
gwsCLI tool to execute listing commands. - Sanitization: Absent. The skill does not specify validation or sanitization for the data returned by the
gwsCLI before the agent processes it.
Audit Metadata