skills/soujava/agent-skills/mermaid/Gen Agent Trust Hub

mermaid

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill defines a set of rules and visual mappings for creating architectural overviews using Mermaid syntax. It operates strictly within the context of text generation and documentation.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes external architectural descriptions.
  • Ingestion points: User-provided architecture names, component lists, and subsystem descriptions (SKILL.md).
  • Boundary markers: Instructions mandate wrapping output in fenced mermaid code blocks (SKILL.md).
  • Capability inventory: The agent is instructed to write the generated diagram into a target file like README.md (SKILL.md).
  • Sanitization: No explicit validation or escaping of user-supplied component names is defined before they are interpolated into the Mermaid diagram structure.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 01:15 PM