o11y-data

What problem does it solve?

This skill eliminates the friction of accessing fragmented observability data across TiDB Cloud's multiple deploy types (Dedicated, Premium, Starter/Essential, BYOC) and cloud providers, removing the need to manually navigate disjointed APIs, storage systems, and authentication flows to gather metrics, logs, slow queries, and profiles for cluster diagnostics.

Core Features & Use Cases

• Automated Cluster Resolution: Uses a 3-tier lookup (Clinic API → GS API scan → user prompt) to automatically resolve cluster provider, region, and deploy type without exhaustive searching.
• Unified Data Access: Authenticates once for AWS, GCP, or Alibaba Cloud to query Prometheus metrics, Loki real-time logs, cloud storage logs/slowlogs/statements, Delta Lake aggregated data, and pprof profiles.
• One-Command Inspection: Runs pre-built scripts to generate full cluster health reports with anomaly detection, including QPS, latency, error counts, top slow queries, and TiKV disk metrics.
• Use Case: A DBA troubleshooting a high-latency TiDB cluster can use this skill to instantly pull all relevant observability data and identify root causes without switching between multiple tools and dashboards.

Quick Start

Use the o11y-data skill to run a full health inspection for TiDB Cloud cluster 10324983984131567830 and generate a report of any detected anomalies.