doc-coauthoring

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

• Third-party content exposure detected (high risk: 0.70). SKILL.md’s Stage 1 explicitly allows pulling in “Related team discussions or shared documents” via runtime integrations (e.g., Slack/Teams/Drive/MCP) after user mention/consent, which would cause outsider-authored free text from those systems (other users’ messages/docs) to be ingested into the LLM context.