What problem does it solve? Azure cost anomaly alerts tell you that spending spiked but not why, leaving engineers to manually correlate billing data with deployments, autoscaling events, and portal changes across multiple tools. ## Core Features & Use Cases - Two Investigation Modes: Look up the latest anomaly for a specific resource with its full trend history, or scan all resources with anomalies in a given month or date range. - Root Cause Diagnosis: Cross-references GitLab deployment pipelines, Azure Activity Log write operations, AKS node count metrics, and resource tags to classify anomalies as deployment-driven, autoscaler bursts, billing glitches, or configuration changes. - Knowledge Base Writeback: Writes investigation findings back to the CSV knowledge base in GitLab and can comment on linked Jira issues, building an annotated history of recurring anomalies. - Use Case: A weekly anomaly email shows a $544 spike on a production CDN resource group. Run the investigation to drill down to the top resource by meter, check for matching GitLab pipelines, and record the confirmed root cause in the knowledge base. ## Quick Start Ask the assistant to investigate why the cost spiked for a specific Azure resource or resource group, optionally naming the subscription or time period.
