wemake-python-mcp
MCP server providing automated code linting, rule explanations, and configuration templates for wemake-python-styleguide, with structured violation reports and offline rule database.
README
wemake-python-mcp
Model Context Protocol (MCP) server providing automated code linting, rule explanations, and configuration templates for wemake-python-styleguide (WPS).
Features
- Automatic Venv Resolution: Automatically detects project virtualenvs (
.venv,venv,uv) to executeflake8 --select=WPS. - Structured Violation Output: Returns JSON violation reports with file, line, column, rule code, title, category, and message.
- Offline WPS Rule Database: Bundled offline lookup database covering all 263 WPS rules (
WPS000–WPS699). - Live Scraper & Updater: Scrapes official documentation to keep rule definitions and links up to date.
- FastMCP Protocol Support: Implements tools and resources via standard
stdiotransport.
Prerequisites
- Python 3.10+
wemake-python-styleguide(installed in target project virtual environment or system environment)uv(recommended) orpip
Installation
Using uv (Recommended)
Clone the repository and install dependencies into a local virtualenv:
git clone https://github.com/your-org/wemake-python-mcp.git
cd wemake-python-mcp
uv venv
uv pip install -e .
Using pip
pip install -e .
Usage & MCP Client Configuration
CLI Usage
Run the server over stdio transport:
wemake-python-mcp --transport stdio
Check version:
wemake-python-mcp --version
Claude Desktop / Cursor / MCP Client Configuration
Add the following to your MCP client configuration file (e.g., claude_desktop_config.json):
Using uv (Recommended)
{
"mcpServers": {
"wemake-python": {
"command": "uv",
"args": [
"--directory",
"/absolute/path/to/wemake-python-mcp",
"run",
"wemake-python-mcp"
]
}
}
}
Direct Executable
{
"mcpServers": {
"wemake-python": {
"command": "/absolute/path/to/wemake-python-mcp/.venv/bin/wemake-python-mcp",
"args": []
}
}
}
Tools Reference
| Tool Name | Parameters | Description |
|---|---|---|
lint_file |
path: str |
Lints a single .py file using wemake-python-styleguide rules. |
lint_project |
directory: str = "." |
Recursively lints a Python project directory. |
lint_code |
code: str |
Lints a raw Python code string snippet. |
explain_rule |
code: str |
Returns documentation, title, category, and URL for a rule (e.g. WPS432 or 432). |
update_rule_db |
None | Scrapes official readthedocs documentation and updates the local rule database. |
Sample Tool Output (lint_file / lint_code)
{
"success": false,
"violation_count": 1,
"violations": [
{
"file": "example.py",
"line": 1,
"column": 5,
"code": "WPS432",
"title": "Forbid magic numbers",
"category": "Best practices",
"message": "Found magic number: 42"
}
],
"summary": "Found 1 wemake-python-styleguide violation across 1 file."
}
Resources Reference
| Resource URI | Mime Type | Description |
|---|---|---|
wemake://rules |
application/json |
Categorized index of all WPS rule codes and titles. |
wemake://rules/{code} |
text/markdown |
Full Markdown documentation for a specific rule (e.g., wemake://rules/WPS432). |
wemake://config/template |
text/plain |
Recommended setup.cfg configuration template for wemake-python-styleguide. |
Environment Variables
WPS_FLAKE8_PATH: Explicit path toflake8executable. Overrides auto-detection logic.
Development & Testing
Run tests and type checks:
# Run unit tests
uv run pytest
# Type check
uv run mypy src/ tests/
# Ruff linting
uv run ruff check src/ tests/ scripts/
# WPS styleguide compliance check
uv run flake8 src/ tests/ --select=WPS
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。