design-system-generator

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. It ingests untrusted user input—including business names, industry types, and brand personality keywords—and incorporates this data into a structured DESIGN.md file. This file is intended to be machine-readable and acts as the 'single source of truth' for downstream build pipelines like website-rebuild and azerbaijan-website-build. Maliciously crafted input could influence the behavior of those downstream agents.
  • Ingestion points: SKILL.md (Input Requirements table: Business name, industry, brand personality keywords, aesthetic inspiration target).
  • Boundary markers: Absent; user inputs are interpolated into the generated document without delimiters or 'ignore' instructions.
  • Capability inventory: Performs file-write operations (DESIGN.md); references the execution of a local script 05-website-rebuild/scripts/contrast-check.js.
  • Sanitization: No sanitization or validation logic is defined for the user-supplied input strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 08:17 AM
Security Audit — agent-trust-hub — design-system-generator