global-mermaid
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill contains a data ingestion surface where it processes user descriptions of systems to generate diagrams.
- Ingestion points: User input collected in Step 1 regarding system steps and context.
- Boundary markers: None specified in the prompt instructions.
- Capability inventory: Limited to generating Mermaid markdown blocks and HTML/CSS for visualization.
- Sanitization: No explicit input validation or sanitization is defined.
- [SAFE]: No malicious patterns or security risks were identified. The skill functions as a visual formatting template without access to high-privilege tools or sensitive environments.
Audit Metadata