monitoring-dashboard

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

• Third-party content exposure detected (high risk: 0.80). This skill integrates with the cloud-hosted Weights & Biases service and explicitly opens and downloads user-run dashboards and artifacts from wandb.ai (see scripts/launch-monitoring.sh opening https://wandb.ai/..., examples and templates using wandb.use_artifact, wandb.restore(), and artifact.download()), which fetches user-generated, third-party content that the agent would read or load as part of its workflow.