DuckDuckGo MCP Server
# DuckDuckGo MCP Server
A Model Context Protocol (MCP) server that provides web search capabilities through DuckDuckGo with content fetching features.
## Features
- **Web Search**: Search DuckDuckGo with advanced rate limiting and result formatting
- **Content Fetching**: Retrieve and parse webpage content with intelligent text extraction
- **Rate Limiting**: Built-in protection against rate limits
- **LLM-Friendly Output**: Results formatted specifically for large language model consumption
## Docker Build
```bash
# Build the Docker image
docker build -t duckduckgo-mcp-server:latest .
# Run the MCP server
docker run --rm -i duckduckgo-mcp-server:latest
```
## Aliyun Image
The image is also available on Aliyun Container Registry:
```bash
# Pull from Aliyun
docker pull crpi-kdbkl9428esejpsc.cn-hangzhou.personal.cr.aliyuncs.com/li-personal/duckduckgo-mcp-server:latest
# Run the Aliyun image
docker run --rm -i crpi-kdbkl9428esejpsc.cn-hangzhou.personal.cr.aliyuncs.com/li-personal/duckduckgo-mcp-server:latest
```
## MCP Tools
### search
Search DuckDuckGo and return formatted results.
- **query**: Search query string
- **max_results**: Maximum number of results (default: 10)
### fetch_content
Fetch and parse content from a webpage URL.
- **url**: The webpage URL to fetch content from
## Claude Code Integration
Add to Claude Code global configuration:
```bash
claude mcp add --scope user --transport stdio duckduckgo-search -- docker run --rm -i crpi-kdbkl9428esejpsc.cn-hangzhou.personal.cr.aliyuncs.com/li-personal/duckduckgo-mcp-server:latest
```
## Project Structure
```
.
├── Dockerfile # Docker build configuration
├── docker-compose.yml # Docker Compose configuration
├── pyproject.toml # Python project configuration
├── .dockerignore # Docker build ignore rules
└── src/
└── duckduckgo_mcp_server/
├── __init__.py # Package initialization
└── server.py # Main MCP server implementation
```TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: 'fetch_content' retrieves and parses content from a specific webpage URL, while 'search' queries DuckDuckGo for results. There is no overlap in functionality, and an agent can easily differentiate between fetching known content and searching for unknown information.
Both tool names follow a consistent verb-based pattern: 'fetch_content' and 'search'. They use snake_case uniformly, and the verbs ('fetch', 'search') are clear and appropriate for their actions, making the naming predictable and readable.
With only 2 tools, the server feels under-scoped for a DuckDuckGo MCP server. While the tools cover basic web content fetching and search, the domain suggests potential for more operations (e.g., image search, news search, or advanced query parameters), making the count too low for the apparent scope.
The tools cover core functionalities of fetching web content and searching, but there are notable gaps. For a DuckDuckGo server, missing operations like image search, video search, or localized searches could limit agent capabilities, though basic workflows are supported.