comfyui-workflow-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted JSON data from ComfyUI workflows provided by the user, which presents a surface for potential instruction injection. However, this is an inherent risk for the skill's primary utility and is considered safe given its limited capabilities.\n
- Ingestion points: Step 1 reads a user-provided workflow JSON file to extract class types and model names.\n
- Boundary markers: There are no specific instructions for the agent to isolate or ignore instructions that might be embedded within the JSON values.\n
- Capability inventory: The skill is restricted to the 'Read' and 'Write' tools, with no access to search tools or general network operations.\n
- Sanitization: No explicit input validation or sanitization of the JSON structure is required by the instructions.
Audit Metadata