recipe-post-mortem-setup
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external incident data via the [Incident] placeholder and interpolates it into tool commands.\n
- Ingestion points: The [Incident] placeholder in SKILL.md captures external data.\n
- Boundary markers: The instructions do not define boundary markers or delimiters for the external content.\n
- Capability inventory: The skill utilizes gws docs +write, gws calendar +insert, and gws chat +send commands in SKILL.md to interact with external services.\n
- Sanitization: No evidence of validation or sanitization of the incident name before it is used in subsequent operations.
Audit Metadata