MCP
@otoTree
MCP について
概要はまだありません
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"mcp-ototree": {
"command": "uv",
"args": [
"init",
"project",
"-",
"name"
]
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is MCP?
MCP (Model Context Protocol) is a client tool that exposes local MCP servers through Flask, enabling platforms like Coze, Dify, and FastGPT to interact with local resources via self-hosted intranet penetration.
How to use MCP?
Set up the environment with uv, create a project directory, activate a virtual environment, and install dependencies from pyproject.toml. Configure mcp_server.json with server definitions and set environment variables in a .env file. Run python server.py to start the Flask server. Use POST requests to endpoints like /list_tool, /call_tool, /list_server, /list_files, and GET requests to /get_download_link/<secret_key>/<filename>.
Key features of MCP
- Exposes local MCP servers via HTTP API.
- Supports multiple server definitions in a JSON config.
- Provides endpoints to list tools, call tools, and list servers.
- Includes file listing and download link generation.
- Uses secret key authentication for API requests.
- Works with self-signed certificates for local development.
Use cases of MCP
- Integrate local MCP tools (e.g., weather, filesystem) with cloud AI platforms.
- Allow Coze, Dify, or FastGPT to call local Python scripts via API.
- Securely expose file management capabilities to external services.
- Debug and test MCP server interactions with a simple REST interface.
FAQ from MCP
What is required to run MCP?
You need uv installed, a virtual environment, and dependencies from pyproject.toml. A mcp_server.json file defines the servers to expose, and a .env file holds sensitive keys.
How do I configure servers in MCP?
Edit mcp_server.json with entries containing desc, command, and args. Each entry represents an MCP server that will be exposed via the Flask API.
Is authentication supported?
Yes, all API endpoints require a secret_key in the request body (or URL path for downloads). The secret is defined in the .env file.
Can I use MCP in production?
The README recommends using formal SSL certificates from a trusted CA (e.g., Let’s Encrypt) for cloud deployment, as self-signed certificates may cause issues on platforms like Coze.
What transports does MCP support?
MCP uses HTTP POST and GET endpoints via a Flask server. The default transport is REST over HTTP/HTTPS. No other transports (like WebSocket) are mentioned.
「その他」の他のコンテンツ
Inbox Zero AI
elie222The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.
MCP Registry
modelcontextprotocolA community driven registry service for Model Context Protocol (MCP) servers.

Peekaboo MCP – lightning-fast macOS screenshots for AI agents
steipetePeekaboo is a macOS CLI & optional MCP server that enables AI agents to capture screenshots of applications, or the entire system, with optional visual question answering through local or remote AI models.
Blender
ahujasidOpen-source MCP to use Blender with any LLM
Inbox Zero AI MCP
elie222The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.
コメント