clawdlink

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

• Third-party content exposure detected (high risk: 0.80). The skill polls a public relay (RELAY_URL: https://clawdlink-relay.vercel.app in lib/relay.js and requests.fetchFriendRequests/processIncoming in lib/requests.js) and ingests user-generated friend requests and messages, and those message fields (e.g., content.urgency and content.context checked in lib/preferences.js::shouldDeliverNow and processed in heartbeat.js/clawdbot.checkMessages) are used to decide delivery and behavior, so untrusted third-party content can materially influence actions.