web-design-guidelines

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user data including text, files, and external URLs, creating a standard surface for indirect prompt injection.
  • Ingestion points: User input fields materialText, materialFile, and materialUrl defined in references/form-schema.json and SKILL.md.
  • Boundary markers: None present to distinguish user content from instructions.
  • Capability inventory: The scripts/run.py script performs network operations (POST requests) to an external API.
  • Sanitization: Input parameters are passed directly to the API without local sanitization.
  • [COMMAND_EXECUTION]: The skill includes a Python runner (scripts/run.py) designed to be executed by the user to interact with the AI Skills API.
  • [EXTERNAL_DOWNLOADS]: The included script performs network requests to the vendor's domain https://ai-skills.ai for the purpose of executing the AI skill's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 03:31 AM
Security Audit — agent-trust-hub — web-design-guidelines