model-qa-specialist
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze model documentation and raw datasets from external sources, which represents a potential attack surface for instructions embedded in data.
- Ingestion points: Documentation review in 'Scoping & Documentation Review' (Phase 1) and raw population reconstruction in 'Data & Feature Quality Assurance' (Phase 2).
- Boundary markers: Absent; the instructions do not explicitly mandate the use of delimiters or 'ignore' directives when processing external content.
- Capability inventory: The provided Python code snippets perform file system write operations to save analysis charts via
matplotlib. - Sanitization: No specific sanitization, filtering, or validation steps are outlined for the external data or documentation processed by the agent.
Audit Metadata