auditing-compliance-and-provenance

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

• Third-party content exposure detected (high risk: 0.75). The required workflow ingests a “compliance scan” free-text block (e.g., licenses/PII/provenance details) that is not authored by the operating user, so that outsider-provided text can be passed into the agent’s LLM context for analysis (indirect prompt-injection risk via the scan content).