qbr-deck-builder
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted customer data which could be used as a vector for indirect instructions.
- Ingestion points: Raw customer feedback and support history are collected in the intake Phase 0.
- Boundary markers: Slide templates for Slide 4 and Slide 8 lack delimiters or 'ignore' instructions for the embedded data.
- Capability inventory: The skill performs file system write operations to the 'clients/' directory in Phase 3.
- Sanitization: The instructions do not include any validation or escaping mechanisms for external inputs.
Audit Metadata