finance-markets-research

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow ingests and LLM-reads untrusted outsider-authored free text from public filings/disclosure content (e.g., SEC EDGAR primary documents via /sec/filing/sections and full-text via /sec/full-text-search, plus congress disclosures via /congress/report) fetched as part of the skill’s required API calls.