Multi-Model Advisor MCP Server

What does “Multi-Model Advisor” MCP Server do?

The Multi-Model Advisor MCP Server is a Model Context Protocol (MCP) server designed to connect AI assistants with multiple local Ollama models, enabling them to query several models simultaneously and combine their responses. This approach, described as a “council of advisors,” allows AI systems like Claude to synthesize diverse viewpoints from different models, resulting in more comprehensive and nuanced answers to user queries. The server supports assigning different roles or personas to each model, customizing system prompts, and integrates seamlessly with environments like Claude for Desktop. It enhances development workflows by facilitating tasks such as aggregating model opinions, supporting advanced decision-making, and providing richer contextual information from multiple AI sources.

List of Prompts

• ⛔ No explicit prompt templates are documented in the repository or README.

List of Resources

• ⛔ No specific MCP resources are listed in the repository or documentation.

List of Tools

• ⛔ The repository does not provide a direct list of tools in a server.py or similar file, nor are tool interfaces explicitly documented in the README or visible file tree.

Use Cases of this MCP Server

• Aggregated Model Opinions: Developers can use the server to obtain multiple perspectives from different Ollama models on a single question, leading to more balanced and informed decisions.
• Role-based Querying: Assigning different roles or personas to each model enables simulation of various expert viewpoints for scenario analysis or brainstorming.
• System Model Overview: By viewing all available Ollama models on the system, users can select the best combination for their specific use case.
• Collaborative AI Decision-Making: The “council of advisors” approach helps in synthesizing diverse model outputs, which is valuable in complex problem-solving or when consensus is needed.
• Workflow Integration: Seamless integration with Claude for Desktop and other MCP-compatible clients supports enhanced developer productivity and easy access to multi-model insights.

How to set it up

Windsurf

1. Ensure Node.js 16.x or higher is installed.
2. Install and run Ollama, and make sure required models are available.
3. Edit your Windsurf configuration file to add the Multi-Model Advisor MCP Server.
4. Add the following JSON snippet to your mcpServers section:

   {
     "multi-ai-advisor-mcp": {
       "command": "npx",
       "args": ["@YuChenSSR/multi-ai-advisor-mcp@latest"],
       "env": {
         "OLLAMA_HOST": "http://localhost:11434"
       }
     }
   }
   

5. Save the configuration and restart Windsurf.
6. Verify the server is running and accessible.

Claude

1. Install Node.js 16.x or higher.
2. Ensure Ollama is running and required models are pulled.
3. Use Smithery for one-step install:

   npx -y @smithery/cli install @YuChenSSR/multi-ai-advisor-mcp --client claude
   

4. Alternatively, add this block to your Claude MCP configuration:

   {
     "multi-ai-advisor-mcp": {
       "command": "npx",
       "args": ["@YuChenSSR/multi-ai-advisor-mcp@latest"],
       "env": {
         "OLLAMA_HOST": "http://localhost:11434"
       }
     }
   }
   

5. Save and restart Claude, then verify integration.

Cursor

1. Install Node.js and Ollama.
2. Edit Cursor’s MCP server configuration to include:

   {
     "multi-ai-advisor-mcp": {
       "command": "npx",
       "args": ["@YuChenSSR/multi-ai-advisor-mcp@latest"],
       "env": {
         "OLLAMA_HOST": "http://localhost:11434"
       }
     }
   }
   

3. Save configuration, restart Cursor, and verify MCP availability.

Cline

1. Ensure prerequisites: Node.js, Ollama, required models.
2. Locate and edit Cline’s MCP configuration file.
3. Add:

   {
     "multi-ai-advisor-mcp": {
       "command": "npx",
       "args": ["@YuChenSSR/multi-ai-advisor-mcp@latest"],
       "env": {
         "OLLAMA_HOST": "http://localhost:11434"
       }
     }
   }
   

4. Save, restart Cline, and confirm MCP is working.

Securing API Keys

To secure API keys or sensitive environment variables, use the env field in your configuration:

{
  "multi-ai-advisor-mcp": {
    "command": "npx",
    "args": ["@YuChenSSR/multi-ai-advisor-mcp@latest"],
    "env": {
      "OLLAMA_HOST": "http://localhost:11434",
      "MY_SECRET_API_KEY": "${MY_SECRET_API_KEY}"
    }
  }
}

Set environment variables in your OS or CI/CD pipeline to avoid hardcoding secrets.

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "multi-ai-advisor-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “multi-ai-advisor-mcp” to the actual name of your MCP server and replace the URL with your own MCP server URL.

Overview

Our opinion

The Multi-Model Advisor MCP is well-documented for setup and provides a unique “council of advisors” approach, but lacks transparency on prompts, resources, and tools. Its value is high for multi-model decision workflows, though more technical detail would improve it. I would rate this MCP a 6/10 based on the two tables, as it covers the basics and offers a compelling use case, but lacks depth in technical documentation.

MCP Score