WebSearch

MCP server for agent web search
  • python

15

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "howe829-websearch-mcp-server": {
      "command": "python",
      "args": [
        "server.py"
      ],
      "env": {
        "CC": "us",
        "HOST": "127.0.0.1",
        "PORT": "8002",
        "LANGUAGE": "en",
        "IMPERSONATE": "edge",
        "SERVER_MODE": "streamable-http",
        "BING_SEARCH_BASE_URL": "https://www.bing.com"
      }
    }
  }
}

WebSearch MCP Server is a lightweight MCP server that lets you perform web searches, convert URL content to Markdown, retrieve WeChat official account articles, and search GitHub. It’s useful when you want a single local service to power search-related features in your MCP client or automation workflows.

How to use

Start the server locally using your preferred approach, then connect with an MCP client to perform actions such as web searches, URL-to-Markdown conversions, and GitHub searches. You can run the server from your development environment and keep it running to handle multiple requests over time.

Common usage patterns include: start the server, set the environment configuration as needed, then issue MCP requests to perform searches or content conversions. The server exposes several capabilities that you can invoke from your MCP client, including performing free web searches with multiple providers, converting URL content to Markdown, retrieving WeChat official account articles, and searching GitHub.

How to install

Prerequisites: Python 3.12 or higher. A dependency management tool such as uv (recommended) or pip.

Step 1: Clone the project and navigate into the directory.

git clone https://github.com/Howe829/websearch-mcp-server.git
cd websearch-mcp-server

Step 2: Install dependencies. You can use uv to create and synchronize a virtual environment, or install via pip.

uv venv && uv sync

Or, using pip:

pip install -r requirements.txt

Step 3: Create and configure environment variables. Copy the example and customize as needed.

cp .env.example .env

Edit the .env file to set your preferences, for example:

BING_SEARCH_BASE_URL="https://www.bing.com"
LANGUAGE="en"
CC="us"
IMPERSONATE="edge"
HOST=127.0.0.1
PORT=8002
SERVER_MODE=streamable-http

Step 4: Start the server. You can start it using uv or directly with Python.

uv run python server.py

Or, activate the virtual environment and run the server with Python:

source .venv/bin/activate
python server.py

Additional content

Configuration notes: The server relies on several environment variables to tailor behavior, such as which search providers to use and how to format content. The recommended defaults are suitable for testing, but you should adjust them for production use.

Security and maintenance: Keep dependencies up to date and run the server in a controlled environment. When deploying, consider restricting access to localhost or implementing authentication on your MCP client.

Troubleshooting tips: If the server fails to start, verify that your Python version is at least 3.12, ensure that the virtual environment is activated (if used), and confirm that the .env file is present and correctly configured.

Available tools

web_search

Performs web searches using multiple providers (Bing, Baidu, Google with browser_use) and returns structured results.

markdown_convert

Converts the content of a provided URL into Markdown format for easy integration.

wx_articles

Retrieves content from WeChat official account articles for display or processing.

github_search

Searches GitHub for repositories, users, issues, and more.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational