mermaid-state-diagram
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE] (SAFE): The skill consists entirely of markdown instructions and reference materials. No scripts, binaries, or other executable code are present.
- [DATA_EXPOSURE] (SAFE): No sensitive file paths, environment variables, or hardcoded credentials were found.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes untrusted input (code and specifications) to generate diagrams. While this is an attack surface, the skill lacks any capabilities (such as network access or command execution) to act on malicious instructions. Evidence: (1) Ingestion points: Input fields 'Code' and 'Specs' in SKILL.md. (2) Boundary markers: Fixed output structure for diagrams and summaries. (3) Capability inventory: No scripts or tools present. (4) Sanitization: None.
- [EXTERNAL_DOWNLOADS] (SAFE): No external dependencies or remote script executions were detected.
Audit Metadata