azure-ai-projects-java

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

• Third-party content exposure detected (high risk: 0.80). The SDK examples and APIs (e.g., DatasetsClient.createOrUpdateVersion and FileDatasetVersion.setDataUri showing a dataUri like "https://...") permit ingesting content from arbitrary public URLs, so the agent can fetch and process untrusted third‑party content that could carry indirect prompt injections.