chaindesk

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

• Third-party content exposure detected (high risk: 0.90). The skill's SKILL.md shows actions that ingest and read untrusted third-party content—e.g., "Get Conversation Messages" (Discord/Telegram/email user messages) and "Create Web Page Datasource"/"Create Web Site Datasource" (fetching arbitrary website URLs)—which the agent is expected to read and which can influence subsequent queries/actions.