design-md

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime path for non-bundled brands uses webfetch to fetch a third-party DESIGN.md from raw.githubusercontent.com/voltagent/awesome-design-md/..., and that fetched free-text markdown is then read by the agent and written into ./DESIGN.md for subsequent LLM context ingestion (outsider-authored public web content).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill fetches DESIGN.md at runtime from https://raw.githubusercontent.com/voltagent/awesome-design-md/main/design-md/{slug}/DESIGN.md and injects that fetched design document into the agent's working files/context to drive prompts and UI generation, so remote content directly controls agent instructions.