Run Locally
For developers who need to run the MCP server locally for development, debugging, or custom modifications.
When to Run Locally
Consider running locally if you need:
- Development and debugging of the MCP server itself
- Testing unreleased features or custom modifications
- Offline work in restricted network environments
- Contributing to the mcp-klever-vm repository
- Custom knowledge entries with Redis persistent storage
Prerequisites
Before configuring your IDE, clone and build the MCP server:
1. Clone the Repository
git clone https://github.com/klever-io/mcp-klever-vm.git
cd mcp-klever-vm
2. Install Dependencies and Build
npm install
npm run build
This creates the dist/index.js file that you'll reference in your IDE configuration.
IDE Configuration
Claude Code
Configure MCP Server
Create or edit your mcp.json file:
{
"mcpServers": {
"klever-vm": {
"command": "node",
"args": [
"--experimental-vm-modules",
"dist/index.js"
],
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "memory"
}
}
}
}
Note: This uses in-memory storage (knowledge resets on restart). For persistent custom knowledge entries, see Storage Options below to configure Redis.
Start Claude Code
claude --mcp-config ./mcp.json
Visual Studio Code
Configure MCP Server
Create a .vscode/mcp.json file at the root of your project:
{
"inputs": [],
"servers": {
"klever-vm": {
"type": "stdio",
"command": "node",
"args": [
"--experimental-vm-modules",
"/absolute/path/to/mcp-klever-vm/dist/index.js"
],
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "memory"
}
}
}
}
Important: Update the path in args to point to your absolute path to mcp-klever-vm/dist/index.js.
Start MCP Server
After configuring mcp.json, VS Code will show an option to run the MCP server directly. You can:
- Click to start the server and view output within VS Code
- Or restart VS Code to auto-discover and start your MCP server
Claude Desktop
Configure
Edit your Claude Desktop config file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"klever-vm": {
"command": "node",
"args": [
"--experimental-vm-modules",
"/absolute/path/to/mcp-klever-vm/dist/index.js"
],
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "memory"
}
}
}
}
Restart Claude Desktop after saving the configuration.
Cursor
Configure
Create a .cursor/mcp.json file at the root of your project:
{
"mcpServers": {
"klever-vm": {
"command": "node",
"args": [
"--experimental-vm-modules",
"/absolute/path/to/mcp-klever-vm/dist/index.js"
],
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "memory"
}
}
}
}
Restart Cursor after creating the configuration file.
Storage Options
Memory Storage (Default)
The default configuration uses in-memory storage, which is fast but doesn't persist across restarts.
Configuration:
{
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "memory",
"MEMORY_MAX_SIZE": "10000"
}
}
Redis Storage
For persistent storage and custom knowledge entries, use Redis.
Configuration:
{
"env": {
"MODE": "mcp",
"STORAGE_TYPE": "redis",
"REDIS_URL": "redis://localhost:6379"
}
}
Setup Steps:
-
Install Redis: Follow the official Redis installation guide
-
Verify Redis is running:
redis-cli ping
# Should return: PONG
- Ingest knowledge base:
cd mcp-klever-vm
npm run ingest
-
Update your MCP configuration with Redis settings (as shown above)
-
Restart your IDE/Claude Code
Development Commands
# Install dependencies
npm install
# Build
npm run build
# Development mode with hot-reload
npm run dev:mcp
# Run tests
npm test
# Lint and format
npm run lint
npm run format
# Ingest knowledge into Redis
npm run ingest
Next Steps
- Learn about MCP Functions: View available tools and capabilities
- Try MCP Prompts: Use guided workflows
- Explore Knowledge Base: Browse curated entries
- See Workflow Example: Complete development workflow
Important Security Note: Always review MCP server configuration before running - MCP servers can execute code on your machine.