data-model-creation
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-supplied business requirement descriptions to generate Mermaid classDiagram code, creating a surface for indirect prompt injection. A malicious user could provide descriptions containing instructions to influence the agent's behavior or manipulate database operations.
- Ingestion points: User business requirements analyzed during the Business Analysis Phase in SKILL.md.
- Boundary markers: The prompt instructions do not specify the use of delimiters or 'ignore' instructions for the content of the user's requirement descriptions.
- Capability inventory: The skill documentation references the ability to use 'relational-database-tool' and perform 'executeWriteSQL' operations.
- Sanitization: There are no instructions provided to validate, filter, or sanitize user input before it is incorporated into the modeling logic.
Audit Metadata