weights-and-biases

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

• Third-party content exposure detected (high risk: 0.80). The skill's artifacts workflow (references/artifacts.md) explicitly shows adding and using external references like artifact.add_reference('https://example.com/data.zip') and run.use_artifact(...)/artifact.download() with subsequent code that reads CSVs and model files, which means the agent fetches and ingests arbitrary HTTP/S3/GCS-hosted third‑party data that can influence training and runtime decisions.