plantuml
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes standard architectural libraries hosted on GitHub for C4 model diagrams (e.g., from the plantuml-stdlib repository). These references are documented neutrally and adhere to standard industry practices for high-quality diagram generation.
- [PROMPT_INJECTION]: The skill defines a clear operational boundary for processing external user requests and project source code. It mitigates indirect injection risks by instructing the agent to use explicit boundary markers (@startuml/@enduml) and provides a validation checklist to ensure no unintended instructions or sensitive metadata are included in the final output.
- [DATA_EXFILTRATION]: No exfiltration vectors or persistence mechanisms were detected. The skill focuses on documentation and provides clear rules against the exposure of credentials, machine-specific paths, or secrets within the generated PlantUML source code.
Audit Metadata