skills.rest
alex-quan-001
alex-quan-001COMMUNITY
Explore all skills by this author →
GitHub

tidb-perf-diagnosis

Diagnose TiDB Cloud cluster performance issues using a triage-then-read pattern.

Updated Jun 3, 2026
Search Tags:#performance#troubleshooting#database#metrics#observability#diagnosis#tidb
One-click install
npx skills add https://github.com/alex-quan-001/nutshell-skills --skill tidb-perf-diagnosis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tidb-perf-diagnosis
Source: https://github.com/alex-quan-001/nutshell-skills/tree/main/skills/diagnosis/tidb-perf-diagnosis
Command: npx skills add https://github.com/alex-quan-001/nutshell-skills --skill tidb-perf-diagnosis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill eliminates guesswork when TiDB Cloud clusters experience latency spikes, OOM, hotspots, slow queries, lock contention, or scaling issues. It provides a structured triage-then-read methodology to classify problems and load only the relevant diagnostic knowledge.

Core Features & Use Cases

  • Triage-then-read pattern: Classify the performance problem first, then load only the relevant knowledge files from the knowledge directory.
  • Multi-architecture support: Covers Dedicated, Starter/Essential, Premium, and BYOC architectures with product-line-specific metrics and procedures.
  • Holistic cross-validation: Mandates TopSQL cross-validation for CPU hotspots and connects metrics with slow queries, logs, SQL statements, and conf profiling.
  • Specialized routing: Automatically routes TiCDC health incidents and transaction-specific oncall issues to dedicated diagnosis skills.

Quick Start

Use the tidb-perf-diagnosis skill to diagnose the current TiDB Cloud performance issue by first classifying the primary symptom and then loading the relevant knowledge file for the cluster type.

Frequently Asked Questions about tidb-perf-diagnosis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I diagnose TiDB Cloud latency spikes and slow queries?▼

Diagnose TiDB Cloud latency spikes and slow queries by applying a triage-then-read pattern that classifies the primary symptom and loads relevant diagnostic knowledge files for your cluster architecture.

What is the best way to troubleshoot TiDB OOM and hotspots?▼

Troubleshoot TiDB OOM and hotspots using a structured methodology that mandates TopSQL cross-validation for CPU hotspot attribution and connects metrics with slow query logs.

Does TiDB performance diagnosis support Starter, Premium, and BYOC architectures?▼

Yes, TiDB performance diagnosis supports Dedicated, Starter/Essential, Premium, and BYOC architectures by applying product-line-specific metrics and diagnostic procedures tailored to each environment.

How do I identify the root cause of TiDB lock contention and scaling problems?▼

Identify the root cause of TiDB lock contention and scaling problems by cross-validating system metrics with SQL statements and slow query logs to trace issues to specific queries.

When should I use a specialized diagnosis skill for TiDB incidents?▼

Use a specialized diagnosis skill for TiDB incidents when the performance issue is identified as a TiCDC health incident or a transaction-specific oncall issue, which are automatically routed to dedicated skills.

Related Skills

Explore More Data & Analytics Skills→
alex-quan-001alex-quan-001

o11y-data

Access TiDB Cloud observability data across metrics, logs, and profiles.

Community
Advanced
randoneeringrandoneering

postgresql

Analyze PostgreSQL configurations, queries, and health metrics for optimization.

Community
Advanced
alex-quan-001alex-quan-001

oncall-toolkit

Automate TiDB Cloud oncall ticket management across Jira and Pingkai platforms.

Community
Advanced
alex-quan-001alex-quan-001

o11y-metrics-api

Query TiDB Cloud observability metrics via a Prometheus-compatible API.

Community
Advanced