mermaid-generator
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill requires the agent to verify all relationships using source evidence and to identify inferences, which acts as a safeguard against malicious or misleading content in analyzed files.
- [INDIRECT_PROMPT_INJECTION]: The skill reads and processes repository files. 1. Ingestion points: Step 3 (Extract observed structure) reads file content. 2. Boundary markers: Absent. 3. Capability inventory: File search and read tools. 4. Sanitization: Absent, but the skill requires strict Mermaid syntax and evidence-based extraction which mitigates potential injection risks.
Audit Metadata