audit

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text from the audited target website is fetched at runtime in the shared fetch phase (tools/audit/fetch.mjs / analyze.mjs) and then passed into each dispatched sub-agent as “fetched HTML for the relevant pages,” which the agents embed verbatim into their LLM context for analysis (untrusted page content can include arbitrary text).