# Tingyun / 听云 MCP

> Tingyun / 听云 connects your AI client to a complete Application Performance Monitoring (APM) system. This MCP lets you talk to your entire performance stack—applications, alerts, metrics, and dependencies—using natural language. Instead of logging into multiple dashboards, your agent instantly lists monitored apps, checks real-time health summaries, and retrieves specific metric data points just by asking a question. It turns complex site reliability work into a simple conversation.

## Overview
- **Category:** developer-tools
- **Price:** Free
- **Tags:** apm, performance-metrics, incident-response, application-monitoring, real-time-alerts, digital-experience

## Description

Tingyun / 听云 gives your AI client control over your whole digital performance stack. You don't need to navigate the monitoring console; you just tell your agent what you want to know, and it handles the rest. Need to know why latency spiked on checkout? Ask your agent. It can instantly list monitored applications for you. Want to check if a database connection is slowing things down? Just ask for dependencies. The tool lets you browse active alerts immediately or query specific metric data points to find anomalies. Everything from external service calls to frontend user experience metrics (RUM) gets organized into simple, conversational answers. By connecting this MCP via Vinkius, your agent acts like a real-time Site Reliability Engineer on demand, keeping your system performance accurate and responsive without you ever leaving your chat window.

## Tools

### get_account_info
Retrieves general metadata about the Tingyun account setup.

### get_app_summary
Provides a high-level performance summary for a specific application.

### get_metrics
Allows querying precise, customizable metric data based on timeframes and dimensions.

### list_alerts
Shows all currently active performance alerts and their associated policies.

### list_app_instances
Lists every running instance of an application to identify geographic or deployment issues.

### list_applications
Retrieves a full list of all Application Performance Monitoring (APM) applications being monitored.

### list_browser_apps
Lists the Real User Monitoring (RUM) applications to audit frontend performance data.

### list_databases
Retrieves a list of all databases monitored for connectivity and query performance issues.

### list_external_services
Lists every external service call an application makes, tracking latency and failure rates.

### list_alert_policies
Shows the predefined rules used to trigger performance alerts within the system.

## Prompt Examples

**Prompt:** 
```
List all applications monitored by Tingyun.
```

**Response:** 
```
I've retrieved your monitored applications. You have 12 APM applications, including 'Checkout Service', 'User API', and 'Auth Gateway'. Which one would you like to check the summary for?
```

**Prompt:** 
```
Show me the performance summary for application ID 12345.
```

**Response:** 
```
I've retrieved the summary for app 12345. Average response time is 150ms, error rate is 0.05%, and Apdex score is 0.95. Would you like to see the database dependency list?
```

**Prompt:** 
```
Check for any critical alerts in Tingyun from today.
```

**Response:** 
```
I've checked for alerts. There is 1 Critical alert for 'High Latency' on the 'Payment Service' and 2 Warnings regarding disk usage. Should I retrieve the alert detail for the high latency incident?
```

## Capabilities

### List monitored applications
Retrieves the names and general health status of all applications being tracked by Tingyun.

### Get application health summaries
Pulls a quick performance report, including average response time and error rates, for a specific service.

### Check active alerts
Lists all current performance warnings or critical incidents that require immediate attention.

### Query system metrics
Fetches precise, historical data points for any measured metric you specify (e.g., CPU usage, latency).

### Audit service dependencies
Lists all connected databases and external services that an application relies on.

### Review user experience data
Browses Real User Monitoring (RUM) applications to audit how the frontend performs for actual end-users.

## Use Cases

### Investigating a sudden spike in checkout errors.
The agent first uses `list_applications` to confirm 'Checkout Service' is monitored. Next, it calls `get_app_summary`. The summary shows the error rate jumped 10x yesterday. You then ask for dependencies, and the tool reveals that one external service call added latency, identifying the root cause instantly.

### Auditing compliance for system health.
An engineering manager needs to know if all core services are monitored properly. The agent uses `list_applications` and then calls `list_alert_policies`. This provides a comprehensive, auditable list of what is covered by monitoring rules.

### Diagnosing slow frontend performance for international users.
The team suspects the mobile experience is poor. Instead of manually checking browser console logs, the agent uses `list_browser_apps` to pull Real User Monitoring data and compare global performance metrics.

### Mapping a complex system's dependencies.
A new developer needs to understand how 'User API' works. They ask the agent, which uses `list_databases` and `list_external_services`, to generate a map of all critical connections, preventing accidental breakage.

## Benefits

- Eliminate dashboard hopping. Instead of opening five tabs to check the 'Checkout Service' summary, you ask your agent directly for the `get_app_summary` and get a single, unified answer.
- Stop guessing where performance issues come from. You can use `list_external_services` to see exactly which third-party API call is causing latency spikes, cutting down debugging time dramatically.
- Stay ahead of downtime with instant awareness. The agent checks for critical alerts using `list_alerts`, giving you immediate notification on 'High Latency' incidents before users complain.
- Understand the full scope of an app. You don't just get a summary; you can run `list_databases` and `list_applications` to map out every dependency, helping pinpoint system bottlenecks.
- Focus on user reality. By using `list_browser_apps`, your agent doesn't just tell you the API is slow; it shows you how the actual customer sees it across different browsers.

## How It Works

The bottom line is you get instant visibility into complex system health using nothing but plain conversation.

1. Subscribe to this MCP and provide your Tingyun API Key and Secret Key.
2. Connect your preferred AI client (like Claude or Cursor) to Vinkius, giving it permission to access the data.
3. Ask a natural language question, such as 'What is the performance summary for the user service?' Your agent translates that request into the necessary tool calls and returns the live data.

## Frequently Asked Questions

**How does Tingyun / 听云 MCP help with performance audits?**
It lets you run comprehensive audits by listing all monitored applications and services. You can use `list_applications` combined with checking application summaries to ensure every part of your stack is accounted for.

**Can I track frontend issues using Tingyun / 听云 MCP?**
Yes, you can check Real User Monitoring (RUM) data by listing browser apps. Using `list_browser_apps` lets your agent audit how the actual end-user views your application on different devices.

**What if I need to know about database changes?**
You can use the MCP's tools to list all monitored databases. This ensures you have a complete inventory and visibility into any potential connectivity or query slowdowns.

**Is Tingyun / 听云 MCP good for SRE teams?**
Absolutely. It’s built for incident response, allowing your agent to list active alerts and quickly retrieve detailed metric data points when an issue arises.