docs/netdata-ai/mcp/mcp-clients/jetbrains-ides.md
Configure JetBrains IDEs to access your Netdata infrastructure through MCP.
JetBrains AI Assistant currently communicates with MCP servers over stdio only (https://www.jetbrains.com/help/ai-assistant/mcp.html).
| Transport | Support | Netdata Version | Notes |
|---|---|---|---|
stdio (via nd-mcp) | ✅ Fully Supported | v2.6.0+ | Launches bridge as subprocess |
stdio (via npx mcp-remote) | ✅ Fully Supported | v2.7.2+ | Wrap remote HTTP/SSE in stdio |
| Streamable HTTP / SSE | ❌ Not Supported | - | Use a stdio launcher |
| WebSocket | ❌ Not Supported | - | Accessible only through nd-mcp |
JetBrains documents a “workaround for remote servers” that relies on launching a stdio wrapper. Native HTTP/SSE support is not available yet.
nd-mcp bridge - Required for Netdata versions that only expose WebSocket (v2.6.0–v2.7.1)npx mcp-remote@latest - Optional wrapper that exposes Netdata HTTP/SSE as stdio (useful for v2.7.2+)export ND_MCP_BEARER_TOKEN="$(cat /var/lib/netdata/mcp_dev_preview_api_key)"
:::note MCP support in JetBrains IDEs may require additional plugins or configuration. Check the plugin documentation for the latest setup instructions. :::
Connect to your entire Netdata Cloud infrastructure through a single endpoint — no local Netdata setup or firewall changes needed.
Prerequisites:
scope:mcp
(create one)Since JetBrains AI Assistant only supports stdio, use npx mcp-remote as a bridge:
AI Assistant Settings:
{
"name": "netdata-cloud",
"command": "npx",
"args": [
"mcp-remote@latest",
"https://app.netdata.cloud/api/v1/mcp",
"--header",
"Authorization: Bearer YOUR_NETDATA_CLOUD_API_TOKEN"
]
}
Replace YOUR_NETDATA_CLOUD_API_TOKEN with your
Netdata Cloud API token (must have scope:mcp).
For more details, see
Netdata Cloud MCP.
The following methods connect directly to a Netdata Agent or Parent on your network.
AI Assistant Settings:
{
"name": "netdata",
"command": "/usr/sbin/nd-mcp",
"args": [
"ws://YOUR_NETDATA_IP:19999/mcp"
]
}
External Tools (if AI Assistant doesn't support MCP directly):
/usr/sbin/nd-mcpws://YOUR_NETDATA_IP:19999/mcpFor detailed options and troubleshooting, see Using MCP Remote Client. JetBrains still launches this command over stdio; mcp-remote converts the remote HTTP/SSE session into the format AI Assistant understands.
AI Assistant Settings:
{
"name": "netdata",
"command": "npx",
"args": [
"mcp-remote@latest",
"--http",
"http://YOUR_NETDATA_IP:19999/mcp",
"--allow-http",
"--header",
"Authorization: Bearer NETDATA_MCP_API_KEY"
]
}
External Tools:
npxmcp-remote@latest --http http://YOUR_NETDATA_IP:19999/mcp --allow-http --header "Authorization: Bearer NETDATA_MCP_API_KEY"Replace in all examples:
/usr/sbin/nd-mcp - With your actual nd-mcp path (nd-mcp method only)YOUR_NETDATA_IP - IP address or hostname of your Netdata Agent/ParentNETDATA_MCP_API_KEY - Your Netdata MCP API keyND_MCP_BEARER_TOKEN - Export with your API key before launching the IDE (nd-mcp method only)Monitor JVM applications:
// Ask AI Assistant about production performance
"What's the memory usage of our Java services?"
"Show me GC patterns in the last hour"
"Are there any thread pool issues?"
Debug Python applications:
# Ask: What's the CPU usage when this function runs in production?
def process_data():
pass
# Ask: Show me memory patterns for the Python workers
Monitor Node.js applications:
// Ask: What's the event loop latency?
// Ask: Show me API endpoint response times
// Ask: Any memory leaks in the Node processes?
Analyze database performance:
-- Ask: Show me database query latency
-- Ask: What's the connection pool usage?
-- Ask: Any slow queries in the last hour?
Add infrastructure context to your code:
@NetdataMonitor("cpu.usage > 80%")
public void resourceIntensiveMethod() {
// AI Assistant can show real-time metrics
}
While debugging:
Combine IDE profiler with Netdata metrics:
"Show me what happened at 14:32 when the error occurred"
"What were the system resources during the last OutOfMemory error?"
"Find correlated metrics during the last service degradation"
"What's the resource usage trend for this service?"
"Project memory needs based on current growth"
"When will we need to scale based on current patterns?"
If official MCP support is limited, consider:
Search marketplace for:
Create a simple plugin that bridges JetBrains with Netdata:
curl http://YOUR_NETDATA_IP:19999/api/v3/info