arc42-section-03
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The generated PlantUML templates reference the C4-PlantUML library from a well-known repository on GitHub (plantuml-stdlib/C4-PlantUML). This is standard behavior for rendering C4 architecture diagrams and the source is considered a well-known service.
- [COMMAND_EXECUTION]: The skill instructs the agent to write diagram source and documentation to local files (e.g., docs/diagrams/context-business.puml). This is the intended primary purpose of the skill and does not involve unauthorized command execution or privilege escalation.
- [PROMPT_INJECTION]: The skill accepts user-provided input regarding system names and actors to populate documentation templates. While this represents an indirect prompt injection surface, the risk is inherent to documentation tools and no malicious capability is exposed. * Ingestion points: User responses to system attribute questions in Step 1. * Boundary markers: Absent; user input is interpolated directly into Markdown and PlantUML templates. * Capability inventory: File system write access for documentation generation. * Sanitization: None; input is processed as architectural metadata.
Audit Metadata