interactive-flowchart-builder
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data to populate the generated interactive flowchart files, creating a potential surface for indirect injection.
- Ingestion points: User-provided steps, tool names, and descriptions provided during the interactive configuration phase.
- Boundary markers: A mandatory verification step is included where the user must explicitly approve the flowchart table before the agent generates any files.
- Capability inventory: The skill uses standard file-writing capabilities to create index.html, tools.js, and workflow.js in the local directory.
- Sanitization: The instructions do not explicitly require sanitization of user-provided strings before they are interpolated into the generated code, which is a common minor surface in code-generation skills.
Audit Metadata