developer-docs-technical-research

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

• Third-party content exposure detected (high risk: 0.85). The required workflow includes “Interview Targeted Sources” (SME interviews) and “Convert verbal answers into written notes,” which at runtime ingests outsider-authored free-form text from engineers/PMs/QA/support/field/cust​omers into the agent’s LLM context.