personize-agent-workspace

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

• Potentially malicious external URL detected (high risk: 1.00). The skill requires agents to call the runtime governance endpoint which returns prompt/instruction content (the shared-workspace-protocol) — e.g. https://agent.personize.ai/api/v1/ai/smart-guidelines — so fetched content from that URL directly controls agent behavior at runtime.