Bing Search MCP

How to Install and Use Bing Search MCP

If you want to add powerful web search and content reading to your project, Bing Search MCP is a good choice. This tool calls the Bing Search API, retrieves search results, reads the content of links, and converts them into a format easy for language models to understand.

Here is a simple guide on how to install and start using Bing Search MCP.

Before You Start: Prepare Your Environment

To get Bing Search MCP working, you first need to set up some environment variables. These are secret keys that the tool needs to access Bing Search and process content.

You will need:

• BING_SEARCH_API_KEY: This key lets Bing Search MCP use the Bing Search API.
• ASSISTANT__AZURE_OPENAI_ENDPOINT: This is needed if you want to process web content further with Azure's OpenAI services.

Make a copy of the .env.example file in the tool’s folder and rename it to .env. Then, open the .env file and add your keys like this:

BING_SEARCH_API_KEY=your_bing_search_api_key_here
ASSISTANT__AZURE_OPENAI_ENDPOINT=your_azure_openai_endpoint_here

Step 1: Install the Bing Search MCP

To install the tool, run the make command inside the mcp-server-bing-search directory. This command creates a virtual environment and installs all needed dependencies.

Open your terminal and type:

make

This step will prepare your system to run Bing Search MCP smoothly.

Step 2: Run the Bing Search MCP Server

Now that the installation is done, you can launch the server. There are two main ways to run it.

1. Standard input/output (stdio) mode — This is the default mode.

Run this command:

uv run mcp-server-bing-search

1. Server-Sent Events (SSE) mode — This allows the server to push data to clients continuously.

Run this command:

uv run mcp-server-bing-search --transport sse --port 6030

If you use SSE mode, you can connect to the server at this address:

http://127.0.0.1:6030/sse

Step 3: Configure Your Client to Use Bing Search MCP

To connect your client to this MCP server, use one of these example configurations depending on how you run the server.

• For stdio mode:

{
  "mcpServers": {
    "mcp-server-bing-search": {
      "command": "uv",
      "args": ["run", "-m", "mcp_server_bing_search.start"]
    }
  }
}

• For SSE mode:

{
  "mcpServers": {
    "mcp-server-bing-search": {
      "command": "http://127.0.0.1:6030/sse",
      "args": []
    }
  }
}

Step 4: Use the Bing Search and Click Tools

Once set up, you can use these two main tools in Bing Search MCP:

• search(query: str) -> str: This tool searches the web for the query you provide, reads the page content, converts it to markdown, and lists useful links.
• click(hashes: list[str]) -> str: This tool lets you pick links using hashes from previous searches and fetches their detailed content, also in markdown.

These tools help you get important information from the web in a format easy for AI models to understand.

---

By following these steps, you will have Bing Search MCP installed and running. It will let your apps search the web and read page content easily with smart results and link handling.