research

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

• Third-party content exposure detected (high risk: 0.90). This skill issues Exa web searches and "WEB SCAN" (e.g., mcp__exa__web_search_exa and Exa search latest developments) to fetch and ingest open web content (tutorials, articles, public sites) which the agent then reads and synthesizes, exposing it to untrusted third-party/user-generated content that could carry indirect prompt injections.