blog-post

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 2 ("Research and gather sources") explicitly tells the agent to "fetch URLs" and "read every source completely," meaning it ingests untrusted public web content that the agent will interpret and use to guide outlining, drafting, quoting, and subsequent actions.