Wikipedia MCP Server
# Wikipedia MCP Server
A Model Context Protocol server that provides capabilities to query Wikipedia. This server enables LLMs to get information from Wikipedia.
## Available Tools
- `search` - Search `keyword` on Wikipedia.
- Required arguments:
- `keyword` (string): The search keyword
- `language` (string): The language
- `fetch` - Fetch Wikipedia page content.
- Required arguments:
- `id` (integer): The pade ID
- `language` (string): The language
## Installation
### Using uv (recommended)
When using [`uv`](https://docs.astral.sh/uv/) no specific installation is needed. We will
use [`uvx`](https://docs.astral.sh/uv/guides/tools/) to directly run *wikipedia-mcp*.
### Using PIP
Alternatively you can install `wikipedia-mcp` via pip:
```bash
pip install wikipedia-mcp-server
```
After installation, you can run it as a script using:
```bash
python -m wikipedia_mcp
```
## Configurations
### Using uvx
```json
{
"mcpServers": {
"wikipedia-mcp": {
"command": "uvx",
"args": [
"wikipedia-mcp-server@latest"
]
}
}
}
```
### Using docker
```json
{
"mcpServers": {
"wikipedia-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"ghcr.io/progamesigner/wikipedia-mcp:latest"
]
}
}
}
```
### Using pip installation
```json
{
"mcpServers": {
"wikipedia-mcp": {
"command": "python",
"args": [
"-m",
"wikipedia_mcp"
]
}
}
}
```
## Build
Docker build:
```bash
docker build -t wikipedia-mcp .
```
TDQS
Scored across 2 tools
The tools have clearly distinct purposes: search finds pages by keyword and returns a list, while fetch retrieves the full content of a specific page by its ID. There is no functional overlap.
Both tool names are single-word verbs ('fetch', 'search') that accurately describe their actions. They follow a consistent simple verb pattern without mixing conventions.
With only 2 tools, the server feels minimal for a Wikipedia MCP server. While the core operations (search and fetch by ID) are covered, 2 tools is on the low side for typical server scopes.
The server lacks several obvious operations, such as fetching a page by title directly (requiring a search then fetch), retrieving random articles, or accessing page summaries. This creates a notable gap in the tool surface.