wkc-workflow-diagram
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of a locally bundled Node.js script (scripts/diagram.mjs) to perform data validation, HTML generation, and local previewing. The execution environment is restricted to Node.js 22 or newer and uses only built-in modules.
- [DYNAMIC_EXECUTION]: The skill uses runtime code generation (new Function) within the bundled Ajv library to compile JSON schemas for validation. This behavior is restricted to processing the local workflow.schema.json and layout.schema.json files and the user's diagram data.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted JSON data from the target project to generate diagrams. 1. Ingestion points: Data is loaded from local workflow.json and layout.json files. 2. Boundary markers: Strict validation is performed against internal schemas (workflow.schema.json and layout.schema.json) before processing. 3. Capability inventory: The skill can execute local scripts, write to a designated diagram directory, and start a local web server on 127.0.0.1. 4. Sanitization: All input is validated against schemas, and the rendering logic includes HTML escaping for titles and other fields to prevent injection in the generated HTML.
- [SAFE]: The skill demonstrates high security standards by implementing a guarded path access function that prevents directory traversal and by explicitly forbidding the use of symbolic links in diagram-related paths.
Audit Metadata