This MCP server wraps the DDGS (Dux Distributed Global Search) API server to expose metasearch results through MCP endpoints. The server aggregates results across multiple backends (text, images, videos, news, books) via the DDGS API and exposes them as MCP endpoints that can be consumed by AI assistants or other MCP-enabled tooling. You can reach the MCP transport at the configured URL (for example, http://localhost:8000/sse) and use the available MCP tools (search_text, search_images, search_news, search_videos, search_books) to obtain aggregated search results. Typical workflows include querying for a topic across multiple backends and receiving a unified set of results suitable for downstream processing by an assistant or workflow.

To use this MCP server within your setup, ensure the DDGS API server is running (the DDGS API exposes endpoints such as /sse for MCP transport, /docs for API docs, and /health for health checks). The MCP client can then route requests to the ddgs-search endpoint to retrieve aggregated results. You can customize backends via the DDGS configuration in your application code or environment, and leverage the explicit tools (text, images, news, videos, books) to perform specialized searches that align with your use case.

Prerequisites:

• Python 3.10+ installed on the system
• Access to install Python packages (pip)

Installation steps (recommended):

1. Create and activate a Python virtual environment (optional but recommended): python3 -m venv venv source venv/bin/activate # on Unix or macOS venv\Scripts\activate # on Windows

2. Install the DDGS package with API support: pip install -U ddgs[api]

3. Run the API server (DDGS) to expose the MCP-compatible endpoints. For example: ddgs api # Start in foreground

   or run in detached mode

   ddgs api -d

4. If you prefer Docker, you can use the repository’s Docker setup as described in the README: git clone https://github.com/deedy5/ddgs && cd ddgs docker-compose up --build

5. Ensure the MCP integration is wired into your MCP runtime by configuring the mcp_config accordingly (see the mcp_config example in this document).

Tips and common considerations:

• The MCP server uses the DDGS API under the hood; ensure the API server is reachable at the configured host/port (default http://localhost:8000).
• Available MCP tools correspond to DDGS search capabilities: search_text, search_images, search_news, search_videos, and search_books. Each tool accepts query parameters aligned with the DDGS API (region, safesearch, timelimit, etc.).
• If you encounter connectivity issues, verify network access between your MCP runtime and the DDGS API server, and confirm that the SSE endpoint (/sse) is exposed.
• For deployment, you can run the API server via Python (pip install ddgs[api]) or use the provided Docker setup; ensure environment variables (if any) required by your deployment are set in your orchestration system.
• The MCP config example uses a single server named ddgs-search; you can adjust the server name or add more servers if you route MCP traffic to multiple DDGS instances.