comfyui-node-frontend
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The documentation defines an architecture for frontend extensions that facilitates the ingestion of untrusted data from the server, creating a surface for indirect prompt injection or cross-site scripting (XSS).
- Ingestion points: SKILL.md describes the use of app.api.addEventListener("executed", ...) to receive arbitrary node outputs from the backend.
- Boundary markers: There are no instructions or patterns provided for delimiting or validating this external content within the extension context.
- Capability inventory: api-reference.md documents high-privilege methods such as fetchApi for authenticated network requests and storeUserData for server-side file manipulation.
- Sanitization: Code examples in SKILL.md demonstrate the use of innerHTML for UI rendering, which can be exploitable if populated with untrusted data from the server without sanitization.
Audit Metadata