ui-blueprint

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill ingests untrusted user content (UI screenshots and project descriptions) to generate design specifications and UI implementation prompts.
  • Ingestion points: Reference image directories (reference-images/) and project idea documents (ideas/*.md) provided by the user in the first step of the workflow.
  • Boundary markers: The skill uses XML-style delimiters (e.g., <context>, <goal>, <format>) in its template assets to separate instructions from injected user data, which helps reduce the risk of the agent misinterpreting data as instructions.
  • Capability inventory: The skill performs file system searches (find), writes documentation to the documents/ folder, and recommends installing standard web development packages via npm install and npx.
  • Sanitization: No explicit sanitization or filtering of the user-provided text or image metadata is described in the workflow instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 12:26 PM
Security Audit — agent-trust-hub — ui-blueprint