ucp-orders-webhooks

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly requires fetching live spec pages from public sites (e.g., ucp.dev/specification and https://ucp.dev/specification/reference and the GitHub conformance repo), so the agent will ingest open third-party web content that can materially influence implementation and runtime verification logic.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to fetch the live UCP spec pages (e.g. https://ucp.dev/specification/reference/) and the conformance test repo (https://github.com/Universal-Commerce-Protocol/conformance) at runtime to obtain the exact order data model and webhook/signature rules, meaning external content would be retrieved and directly determine the agent's implementation instructions.