browser-automation

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md and accompanying examples explicitly instruct the autonomous browser_subagent to navigate to and scrape arbitrary public websites (e.g., examples like https://shop.example.com/products, blog listings, and competitor sites) so the agent will ingest untrusted third-party web content that could contain instructions influencing its actions.