ollama-search-fetch-mcp

An MCP stdio server that exposes Ollama's web_search and web_fetch APIs as tools for any MCP-compatible client (Claude Desktop, Cursor, VS Code Copilot, etc.).

Prerequisites

• Node.js >= 18
• An Ollama API key — get one at ollama.com (required for the hosted search/fetch endpoints)

Installation

npx (no install required)

The easiest way to use this server is via npx:

{
  "mcpServers": {
    "ollama-search-fetch": {
      "command": "npx",
      "args": ["-y", "ollama-search-fetch-mcp"],
      "env": {
        "OLLAMA_API_KEY": "<your-api-key>"
      }
    }
  }
}

Global install

npm install -g ollama-search-fetch-mcp

Then reference the installed binary in your MCP client config:

{
  "mcpServers": {
    "ollama-search-fetch": {
      "command": "ollama-search-fetch-mcp",
      "env": {
        "OLLAMA_API_KEY": "<your-api-key>"
      }
    }
  }
}

From source

git clone https://github.com/rohithmahesh3/ollama-search-fetch-mcp.git
cd ollama-search-fetch-mcp
npm install
npm run build

Then point your MCP client at the built entry point:

{
  "mcpServers": {
    "ollama-search-fetch": {
      "command": "node",
      "args": ["/path/to/ollama-search-fetch-mcp/dist/index.js"],
      "env": {
        "OLLAMA_API_KEY": "<your-api-key>"
      }
    }
  }
}

Configuration

MCP Client Setup

Add the server to your MCP client configuration file. Common locations:

| Client | Config file | |--------|-------------| | Claude Desktop | ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows) | | Cursor | .cursor/mcp.json in your project root | | VS Code Copilot | .vscode/mcp.json in your project root |

Environment Variables

| Variable | Required | Description | |----------|----------|-------------| | OLLAMA_API_KEY | Yes | Your Ollama API key (used as Authorization: Bearer <key>) |

Tools

web_search

Perform a web search using Ollama's hosted search API.

Parameters:

| Name | Type | Required | Default | Description | |------|------|----------|---------|-------------| | query | string | Yes | — | The search query to run | | max_results | integer | No | 3 | Maximum number of results to return (1–10) |

Example response:

{
  "results": [
    {
      "content": "Relevant search result content..."
    }
  ]
}

web_fetch

Fetch the content of a web page for the provided URL.

Parameters:

| Name | Type | Required | Description | |------|------|----------|-------------| | url | string | Yes | The absolute URL to fetch |

Example response:

{
  "title": "Example Page",
  "url": "https://example.com",
  "content": "Page content in markdown format...",
  "links": ["https://example.com/about", "https://example.com/contact"]
}

Development

# Install dependencies
npm install

# Build
npm run build

# Watch mode
npm run dev

# Start the server
npm start

License

MIT