Nia MCP Integration

This guide explains how to set up the Nia MCP server to enhance AI assistants with codebase understanding capabilities.

You need a Nia API key to get started.
For Cursor and Windsurf integrations, you must use your Cursor API key found in the API keys page of your account.

Overview

The Nia Model Context Protocol (MCP) integration enables AI assistants and applications to access Nia’s powerful codebase understanding capabilities through a standardized interface. Using the MCP server, AI assistants like Claude can seamlessly query and retrieve relevant code context from repositories indexed in Nia.

AI Assistants

Enhance AI assistants like Claude with deep codebase understanding

Code Editors

Integrate with AI-powered code editors like Cursor and Windsurf

Custom Applications

Build custom applications with codebase understanding capabilities

Natural Language

Query your entire codebase using natural language

How It Works

The Nia MCP server acts as a bridge between AI assistants and your codebase knowledge:

Setup

The MCP server connects to Nia using your API key

Query

AI assistants send natural language queries about your code

Retrieval

The server retrieves relevant code snippets and context from Nia

Response

Context is returned in a structured format with source references

Installation

You can use the Nia MCP server without installing it using npx:

npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY --transport=stdio

Or install it globally:

npm install -g nia-codebase-mcp

Configuration Options

# Basic usage with stdio transport (default)
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY

# Using SSE transport on custom port
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY --transport=sse --port=3000

# Enable debug mode
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY --debug=true

# Basic usage with stdio transport (default)
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY

# Using SSE transport on custom port
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY --transport=sse --port=3000

# Enable debug mode
npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY --debug=true

# Configure with environment variables
export NIA_API_KEY=your_api_key
export TRANSPORT=stdio
export PORT=3000
export DEBUG=false
export TOOL_TIMEOUT_MS=300000

# Run the server
npx -y nia-codebase-mcp@1.0.1

Option	Description	Default
`--api-key`	Your Nia API key	Required
`--transport`	Connection mode (`stdio` or `sse`)	`stdio`
`--port`	Port for SSE server	3000
`--debug`	Enable debug logging	`false`
`--tool-timeout-ms`	Custom timeout for tool execution	300000ms

Integration with AI Assistants

Configure Claude Desktop

Edit your Claude Desktop configuration at:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add MCP Server Configuration

Add the following to your configuration file:

{
  "mcpServers": {
    "nia-codebase": {
      "command": "npx",
      "args": ["-y", "nia-codebase-mcp@1.0.1", "--api-key=YOUR_NIA_API_KEY"]
    }
  }
}

Restart Claude

Restart Claude Desktop to apply the changes

Configure Claude Desktop

Edit your Claude Desktop configuration at:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add MCP Server Configuration

Add the following to your configuration file:

{
  "mcpServers": {
    "nia-codebase": {
      "command": "npx",
      "args": ["-y", "nia-codebase-mcp@1.0.1", "--api-key=YOUR_NIA_API_KEY"]
    }
  }
}

Restart Claude

Restart Claude Desktop to apply the changes

Open Cursor Settings

Navigate to Settings > Features > MCP Servers

Add New MCP Server

Configure with these settings:

Name: nia-codebase
Type: command
Command: npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY

Note on API Key

Remember to use your Cursor API key, not a standard Nia API key. The Cursor API key can be found in the API keys page of your Cursor account.

Configure MCP Server

For editors that support the Model Context Protocol, configure the MCP server using:

Transport: stdio or SSE
Command: npx -y nia-codebase-mcp@1.0.1 --api-key=YOUR_NIA_API_KEY
Connection Details: For SSE, connect to http://localhost:3000 (or your specified port)

MCP Configuration File Examples

For code editors like Cursor and Windsurf that support MCP, you can use a configuration file to set up the Nia integration.

{
  "Nia": {
    "command": "npx",
    "args": [
      "-y",
      "nia-codebase-mcp",
      "--api-key=YOUR_CURSOR_API_KEY",
      "--transport=stdio"
    ]
  }
}

Replace YOUR_CURSOR_API_KEY with your actual Cursor API key. For Cursor and Windsurf integrations, you must use your Cursor API key specifically, not a standard Nia API key.

Using the MCP Server

Once configured, you can use prompts like these with your AI assistant:

Using the lookup_codebase_context tool, please analyze my project structure and tell me what the main components are.

Technical Details

Tools Available

lookup_codebase_context: Look up context from a codebase indexed in Nia, retrieving relevant code snippets based on user queries

Requirements

Node.js 18 or higher
A valid Nia API key (Cursor API key for Cursor/Windsurf integrations)
A codebase indexed in Nia

Transport Options

stdio: Default transport for most applications
SSE: Server-Sent Events for web and network-based integrations

Performance

Built-in retry logic
Configurable timeout management
Robust error handling

Security & Privacy

The Nia MCP server maintains the same security standards as Nia:

Encrypted Communication

All communication with Nia’s API is encrypted

API Key Security

Your API key is never exposed to the AI assistant

Secure Retrieval

Code context is retrieved securely from your indexed repositories

Privacy Controls

Maintains all privacy controls from your Nia configuration

Get Started

Command Line Interface

Integrations

Overview

AI Assistants

Code Editors

Custom Applications

Natural Language

How It Works

Installation

Configuration Options

Integration with AI Assistants

MCP Configuration File Examples

Using the MCP Server

Technical Details

Tools Available

Requirements

Transport Options

Performance

Security & Privacy

Encrypted Communication

API Key Security

Secure Retrieval

Privacy Controls

Troubleshooting

FAQ

Get Started

Command Line Interface

Integrations

​Overview

AI Assistants

Code Editors

Custom Applications

Natural Language

​How It Works

​Installation

​Configuration Options

​Integration with AI Assistants

​MCP Configuration File Examples

​Using the MCP Server

​Technical Details

Tools Available

Requirements

Transport Options

Performance

​Security & Privacy

Encrypted Communication

API Key Security

Secure Retrieval

Privacy Controls

​Troubleshooting

​FAQ

Overview

How It Works

Installation

Configuration Options

Integration with AI Assistants

MCP Configuration File Examples

Using the MCP Server

Technical Details

Security & Privacy

Troubleshooting

FAQ