c

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "Available Skills" explicitly lists capabilities that fetch and ingest public, user-generated content (e.g., "c-research — Web research + summarization", "c-browser — Headless browser", plus bridges to Slack/Telegram/GitHub), and the Routing instructions require using the most relevant skills, so the agent will read and act on untrusted third-party web/social content.