umple-diagram-generator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (LOW): The skill executes multiple system commands including
umple,dot,mkdir, andcp. These are necessary for its primary function of model compilation and diagram rendering. - [EXTERNAL_DOWNLOADS] (LOW): The script is invoked using
npx -y bun, which may download the Bun runtime from the npm registry at runtime. This is a standard developer workflow but constitutes a remote dependency fetch. - [PROMPT_INJECTION] (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) as it transforms untrusted user requirements into executable Umple code.
- Ingestion points: Natural language requirements provided by the user in the prompt.
- Boundary markers: No specific delimiters or safety warnings are implemented to separate user data from the generation logic.
- Capability inventory: The
umpletool processes the generated.umpfile, which can contain code-based 'actions' or 'guards'. - Sanitization: The skill relies on a manual guardrail instructing the agent to keep actions minimal, rather than technical validation.
Audit Metadata