realtime-sync

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md and referenced docs like references/pubsub-delivery.md and references/ai-stream-orchestration.md) explicitly subscribes to and ingests messages from public pub/sub channels (e.g., ably.channels.get(...); channel.subscribe(...); channel.history(...)) and CRDT updates—user-generated, untrusted content that the agent applies to document state and UI flows, so third-party content can materially influence subsequent actions.