• ISOLATION
  • DATABASE
  • SUMMARY
  • HUMAN

How to troubleshoot application performance with your AI

Connect Tingyun to your AI client to turn telemetry into answers. Once connected, your agent can pull real-time data to find exactly where your code is stalling.

Ask AI about this page

Short answer

How can my AI find application lag?

Your AI uses Tingyun to pull application summaries and specific performance metrics. It compares these against external service calls and database response times to isolate the bottleneck. You'll get a clear breakdown of whether the issue is your code, a third-party API, or a database query.

Where the data goes.

Troubleshooting outcomes.

The AI processes the telemetry to give you these specific results.

ISOLATION

Identify the bottleneck

The AI compares service latency against app performance. You'll know if a specific external API is the culprit.

DATABASE

Spot slow queries

Your agent looks at database performance data. It identifies which specific dependencies are dragging down response times.

SUMMARY

Get a health report

The AI pulls a high-level overview of your application state. This gives you a baseline for current performance levels.

HUMAN

Decide on a fix

The AI presents the evidence, but you decide whether to scale a service, optimize a query, or contact a vendor.

The workflow

What your AI does when the lag occurs.

The AI handles the heavy lifting of data gathering so you can focus on the fix.

  1. Assess the current state

    The AI starts by pulling a high-level view of your application to understand its current status.

    get_app_summary

  2. Check external dependencies

    It looks at every third-party service your app talks to. This helps rule out external API latency.

    list_external_services

  3. Inspect database health

    The agent checks your database connections and performance to see if the storage layer is the problem.

    list_databases

  4. Analyze specific metrics

    Once a suspect is found, the AI pulls granular performance data to confirm the exact point of failure.

    get_metrics

Try it

Copy these to start.

Use these prompts to kick off your investigation.

Starting points

Replace the bracketed text with your specific service or app names.

Tingyun / 听云 Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_LYTPskntXjT3hDeJtLttKdaHELmt1TtQotKDAVf3/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Tingyun / 听云 capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "tingyun-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_LYTPskntXjT3hDeJtLttKdaHELmt1TtQotKDAVf3/mcp"
    }
  }
}
Copy into chat04

  • Show me a summary of the 'payment-gateway' app and tell me if it's running slow.

  • List all external services used by my application and check for high latency.

  • Which databases are currently connected to my production environment?

  • Get the latest performance metrics for my primary database to see if query times have spiked.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect Tingyun once, then ask.

Just follow the link to authorize the connection. Your credentials stay encrypted, and you can immediately start asking your AI about your app's performance.

Connect Tingyun to your AI

FAQ

Troubleshooting with Tingyun

  • 01

    Can the AI fix the slow code for me?

    No. The AI identifies the bottleneck and provides the data, but you must implement the actual code changes or configuration fixes.

  • 02

    Does the AI have permission to delete my databases?

    No. The tools only allow the AI to read summaries, list services, and fetch metrics. It cannot perform write or delete operations.

  • 03

    How does the AI know which service is slow?

    It compares the metrics from your application against the response times of the external services and databases you have listed.

  • 04

    Can I use this with any AI client?

    Yes, as long as your client is MCP-compatible, like Claude, Cursor, or Windsurf.

  • 05

    Will this work for my local development environment?

    The AI can only see the data provided by the Tingyun connection, so it works for any environment you have configured within Tingyun.

  • More questions about Tingyun? See everything the Tingyun Connector can do

Connect Tingyun / 听云 to Claude, Cursor, ChatGPT & more