Web Scout

Web Scout is an MCP server that provides AI-assisted web searching and content summarization. Built as a FastAPI application, it connects DuckDuckGo search capabilities with Google Gemini 2.5 Flash to retrieve online results and process them into structured overviews. Software engineers, researchers, and AI tool users employ Web Scout to grant local language models direct access to real-time information from the web. The server accepts queries and can generate either a concise summary or a detailed analysis based on retrieved source material. It exposes its functionality via standard Model Context Protocol endpoints over HTTP as well as direct REST endpoints, returning structured JSON payloads that include the original query, the chosen operational mode, the AI-generated synthesis, and the total count of sources consulted during processing. By running locally or via Docker containers, Web Scout supplies background research capabilities to desktop AI clients without requiring complex browser automation setups.

Category: Browser & Web Automation

Tags: data-extraction, search, web analysis, web scraping

Visit Web Scout

How to install and configure Web Scout

1. Clone the repository from GitHub and navigate to the project directory. 2. Ensure you have your Google Gemini API key ready. 3. Add your configuration to an MCP client like Claude Desktop by updating your claude_desktop_config.json: json { "mcpServers": { "web-scout": { "command": "python", "args": ["-m", "uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"], "env": { "GEMINI_API_KEY": "your_gemini_api_key_here" } } } } 4. Alternatively, install requirements using pip install -r requirements.txt, set GEMINI_API_KEY, and launch with python main.py or run via Docker Compose with docker-compose up --build to use the /mcp HTTP endpoint.

What you can do with Web Scout

• Performing fast DuckDuckGo web searches directly inside conversational LLM interfaces without manually opening a browser. - Generating concise AI summaries of current news events using Gemini 2.5 Flash from within Claude Desktop. - Requesting detailed multi-source analyses of complex technical questions to extract key facts and findings. - Integrating real-time web retrieval pipelines into autonomous developer agents running in environments like Roo. - Self-hosting a containerized web search service via Docker Compose for local AI tool workflows.

Key facts

• https://github.com/UpperMoon0/Web-Scout-MCP-Server
• Browser & Web Automation, Web Search & Research
• data-extraction, search, web analysis, web scraping

Part of MCP Servers

What can Web Scout do?

Web Scout performs web searches using DuckDuckGo and analyzes the results with Google Gemini 2.5 Flash. It exposes a web search tool that returns either concise summaries or detailed multi-source overviews. It functions both as an MCP server over HTTP and as a standalone FastAPI service.

How do I install Web Scout?

You can install Web Scout by cloning its repository, installing dependencies with pip, and configuring your Gemini API key in an environment variable. Alternatively, you can run the application container using Docker Compose, which builds the service and exposes it on port 8000.

Which MCP clients work with Web Scout?

Web Scout integrates with MCP-compatible clients such as Claude Desktop and Roo. It can be launched directly as a Python process via standard client configuration files or connected over HTTP through its dedicated mcp endpoint when running in a standalone container.

What tools are provided by the Web Scout MCP server?

The server provides the web_search tool. This tool accepts a mandatory search query string and an optional mode parameter, which can be toggled between summary and detailed analysis modes, returning structured JSON containing the processed text and source counts.