create-agent-with-sanity-context

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

• Potentially malicious external URL detected (high risk: 1.00). Yes — the skill fetches runtime instructions/schema from the Sanity MCP endpoint (e.g. https://api.sanity.io/v2026-03-03/context/mcp/:projectId/:dataset/:slug and the /initial-context path) and injects that fetched content directly into the system prompt and tool set, so external content controls agent behavior.