qa

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

• Third-party content exposure detected (high risk: 0.85). At runtime the workflow fetches public origin pages over HTTP for each _source_url (“Fetch the origin page via HTTP”, then “Parse both — origin HTML and WXR content — ... using parseContent()`), so outsider-authored free text/HTML from arbitrary web content becomes LLM-readable in the agent context for comparison/diff.