doc-coauthoring

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs the agent to read user-provided shared documents/links and to pull in context from team channels or other connectors during “Context Gathering,” meaning the agent may ingest outsider-authored free-form text (e.g., content from Slack/Docs/threads) into the LLM context via those integration reads.