plan-canvas

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

• Third-party content exposure detected (high risk: 0.85). The required runtime path is ecc-plan-canvas await <file>, which returns human feedback as JSON (including kind: "chat" freeform text and kind: "annotation" text/snippets) that the agent then reads from stdout and feeds into the LLM context, and that feedback is authored by an outsider (the human reviewer) rather than the operating user.