LOCAL_AI_SETUP.md•1.9 kB
# Local AI MCP Server Configuration
Add this to your Claude Desktop config:
```json
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-filesystem", "/Users", "D:\\", "D:\\arch_mcp\\"],
"env": {}
},
"brave-search": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-brave-search"],
"env": {"BRAVE_API_KEY": "BSA6BeKFEOu8cNP0tj0QXcaVcknsMKe"}
},
"puppeteer": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-puppeteer"],
"env": {}
},
"memory": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-memory"],
"env": {}
},
"architecture-server": {
"command": "node",
"args": ["D:\\arch_mcp\\architecture_server.js"],
"env": {}
},
"cot-server": {
"command": "node",
"args": ["D:\\arch_mcp\\cot_server.js"],
"env": {}
},
"local-ai-server": {
"command": "node",
"args": ["D:\\arch_mcp\\local-ai-server.js"],
"env": {}
}
}
}
```
## Installation
1. Install dependencies:
```bash
cd D:\arch_mcp
npm install
```
2. Ensure Ollama is running:
```bash
ollama serve
```
3. Test the server:
```bash
node local-ai-server.js
```
4. Add to Claude Desktop config and restart
## Available Tools
- **query_local_ai**: Direct queries to local AI models
- **reasoning_assist**: Structured reasoning with specified steps
- **model_list**: List available local models
- **hybrid_analysis**: Combined local+cloud analysis
- **token_efficient_reasoning**: Delegate heavy reasoning to local AI
## Usage Examples
"Use local AI to reason through this complex problem step by step"
"Query my DeepSeek R1 model about this technical question"
"Show me what local models I have available"
"Perform hybrid analysis on this data using reasoning approach"