Frequently Asked Questions about the ddg-mcp MCP Server

Q: What is an MCP Server? A: MCP (Model Context Protocol) is an open protocol that standardizes how applications provide context to LLMs. An MCP server acts as a bridge, allowing AI models to access and interact with external data sources and tools.

Q: What is the ddg-mcp MCP Server? A: The ddg-mcp MCP server provides DuckDuckGo search capabilities through the Model Context Protocol. It allows AI agents to perform web searches, image searches, news searches, video searches, and even chat with DuckDuckGo AI.

Q: What are the prerequisites for installing the ddg-mcp server? A: You need Python 3.9 or higher and either uv (recommended) or pip package manager.

Q: How do I install the ddg-mcp server? A: You can install it from PyPI using uv install ddg-mcp or pip install ddg-mcp. Alternatively, you can install it from source by cloning the repository and running uv install -e . or pip install -e ..

Q: What dependencies are required? A: The server requires the duckduckgo-search package, which is installed automatically when you install ddg-mcp. If you need to install it manually, use uv install duckduckgo-search or pip install duckduckgo-search.

Q: How do I configure the server? A: The server can be configured using command-line arguments. For example, you can specify the transport protocol (--transport stdio for stdio transport) and the host and port for SSE transport (--host 0.0.0.0 --port 3001).

Q: What search parameters can I use? A: You can use common parameters like region, safesearch, timelimit, and max_results for most search types. There are also specific parameters for image and video searches, such as size, color, type_image, layout, and license_image for images, and resolution, duration, and license_videos for videos.

Q: What are the available AI Chat models? A: The server supports several AI models, including gpt-4o-mini, llama-3.3-70b, claude-3-haiku, o3-mini, and mistral-small-3.

Q: How do I use the ddg-text-search tool? A: Use the ddg-text-search tool with the keywords argument to specify the search query. You can also use optional parameters like region, safesearch, timelimit, and max_results.

Q: How do I use the ddg-image-search tool? A: Use the ddg-image-search tool with the keywords argument to search for images. You can also use optional parameters like size, color, type_image, layout, and license_image.

Q: How do I use the ddg-news-search tool? A: Use the ddg-news-search tool with the keywords argument to search for news articles. You can also use optional parameters like region, safesearch, and timelimit.

Q: How do I use the ddg-video-search tool? A: Use the ddg-video-search tool with the keywords argument to search for videos. You can also use optional parameters like resolution, duration, and license_videos.

Q: How do I use the ddg-ai-chat tool? A: Use the ddg-ai-chat tool with the keywords argument to send a message or question to the AI. You can also specify the AI model to use with the model argument.

Q: How do I use the search-results-summary prompt? A: Use the search-results-summary prompt with the query argument to specify the search term. You can also use the style argument to control the detail level (brief/detailed).

Q: How do I configure the server to work with Claude? A: You need to modify the claude_desktop_config.json file in the appropriate directory (MacOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%/Claude/). Detailed configuration examples for both development and published servers are provided in the documentation.

Q: How do I debug the MCP server? A: Debugging MCP servers can be challenging. It’s strongly recommended to use the MCP Inspector for the best debugging experience.

Q: Where can I find more information about UBOS and its AI Agent Development Platform? A: Visit the UBOS website at https://ubos.tech to learn more about their full-stack AI Agent Development Platform and how it can help you orchestrate AI Agents, connect them with your enterprise data, and build custom AI Agents.