conventions-mermaid
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Mermaid CLI (mmdc) to render diagrams and uuidgen to create unique, safe filenames in the /tmp/ directory.
- [EXTERNAL_DOWNLOADS]: The instructions recommend installing @mermaid-js/mermaid-cli from the official NPM registry if the utility is not already present in the environment.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted Mermaid diagram source code as data. While it lacks explicit boundary markers or input sanitization, the capabilities are limited to local diagram rendering and validation, and the shell commands use generated identifiers to mitigate injection risks. Mandatory Evidence Chain: (1) Ingestion Point: User-provided Mermaid diagram source; (2) Boundary Markers: Absent; (3) Capability Inventory: Shell execution (mmdc, uuidgen) and file system writes (/tmp/); (4) Sanitization: Absent.
Audit Metadata