craftsman-agent-build-plans

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

• Third-party content exposure detected (high risk: 0.80). The skill's scripts and SKILL.md call the external OneKey Gateway API at https://agent.deepnlp.org/agent_router and explicitly consume the returned JSON (overall_image, inventory_list, assembly_step_image) to render step-by-step assembly instructions and inventories, so untrusted third‑party responses are ingested and can influence the agent's subsequent behavior.