investment-research-briefs

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly instructs the agent to “use search and extract” from public/third-party sources (e.g., filings, IR pages, regulators, financial media) and to verify claims against those sources, which at runtime can ingest outsider-authored free text into the LLM context via the search/extract steps.