Polars Docs MCP
A FastMCP tool to search and retrieve Polars API documentation with support for multiple transport methods.
README
Polars Docs MCP
A FastMCP tool to search and retrieve Polars API documentation with support for multiple transport methods.
Features
- Automatically discover Polars public components (classes, functions, submodules).
- Search Polars API by component or query string.
- Get current Polars version and package information.
- Returns structured JSON with API signatures and descriptions.
- Support for multiple transport methods: STDIO, Streamable HTTP, and SSE.
- Integrates with
mcpfor seamless LLM-powered workflows.
By leveraging Python's built‑in introspection to reflectively discover every public class, function, and submodule in Polars at runtime, I eliminate the cost, fragility, and maintenance burden of web‑scraping or managing an external documentation database. This approach guarantees 100% up‑to‑date accuracy with every library release, requires no complex text cleaning or embedding pipelines, and avoids the heavy infrastructure overhead of semantic search—making it both simpler and far more efficient for real‑time API lookup.
Usage
1. Claude Desktop Config (Recommended)
{
"mcpServers": {
"polarsapifinder": {
"command": "uv",
"args": [
"--directory",
"/PATH/TO/polars-docs-mcp",
"run",
"polarsdocsfinder.py"
]
}
}
}
2. Manual Execution with Transport Options
STDIO (Default - Best for Claude Desktop)
python polarsdocsfinder.py
# or explicitly:
python polarsdocsfinder.py --transport stdio
Streamable HTTP (Best for Web Deployments)
python polarsdocsfinder.py --transport streamable-http
# With custom settings:
python polarsdocsfinder.py --transport streamable-http --host 0.0.0.0 --port 8080 --path /api/mcp
SSE (For Legacy Client Compatibility)
python polarsdocsfinder.py --transport sse
# With custom settings:
python polarsdocsfinder.py --transport sse --host 0.0.0.0 --port 9000
Command Line Arguments
--transport: Choose transport method (stdio,streamable-http,sse) - Default:stdio--host: Host address for HTTP/SSE transports - Default:127.0.0.1--port: Port number for HTTP/SSE transports - Default:8111--path: URL path for streamable-http transport - Default:/mcp
3. Visual Testing of MCP Server
npx @modelcontextprotocol/inspector uv run polarsdocsfinder.py

Requires Python 3.11+.
Tool Endpoints
get_polars_version(): Get current Polars version and package information.list_polars_components(): List all high-level Polars API components.search_polars_docs(api_refs: list[str] | None, query: str | None, max_results: int = 1000): Search and retrieve API signatures.verify_polars_api(api_ref: str): Verify if a Polars API reference is valid.list_all_modern_data_stacks(): List modern data stacks compatible with Polars.
Transport Methods
STDIO (Default)
- Best for: Claude Desktop, local tools, command-line scripts
- Usage: Direct integration with Claude Desktop configuration
- Communication: Standard input/output streams
Streamable HTTP
- Best for: Web deployments, REST API integration, browser-based clients
- Usage: Run as HTTP service, connect via HTTP requests
- Communication: HTTP POST requests to the specified endpoint
SSE (Server-Sent Events)
- Best for: Compatibility with existing SSE clients, real-time streaming
- Usage: Legacy system integration
- Communication: Server-sent events over HTTP
Examples Snapshots


Testing HTTP Transport
If running with HTTP transport, you can test the server:
# Start server
python polarsdocsfinder.py --transport streamable-http --port 8111
# Test with curl
curl -X POST http://127.0.0.1:8111/mcp \
-H "Content-Type: application/json" \
-d '{"method": "tools/list"}'
License
This project is licensed under the MIT License. See the LICENSE file for details.
Contact
Created by ABC. Report issues or request features at https://github.com/HotTechStack/polars-docs-mcp/issues.
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。