MCP.so
登录

mcp-label-studio: A Label Studio MCP server

@lxkaka

代码仓库

关于 mcp-label-studio: A Label Studio MCP server

mcp server for label studio

#mcp-label-studio#label-studio#data-labeling

基本信息

分类

其他

运行时

python

传输方式

stdio

发布者

lxkaka

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-label-studio": {
      "command": "uv",
      "args": [
        "pip",
        "install",
        "-e",
        "."
      ]
    }
  }
}

工具

15

A list of all projects with their basic information

`project_id` (string): Label Studio project ID

Detailed project information including label configuration

`title` (string): Project title

New project information

`project_id` (string): Project ID

Updated project information

`project_id` (string): Project ID to delete

Deletion status

`project_id` (string): Project ID

Exported annotation data

`project_id` (string): Project ID

Import status

`project_id` (string): Project ID

List of supported export formats

概览

What is mcp-label-studio?

mcp-label-studio is a Model Context Protocol (MCP) server that connects to Label Studio to manage and interact with annotation projects. It provides tools to create, update, delete projects, import tasks from files, and export annotations in multiple formats, enabling automated data labeling workflows for AI practitioners.

How to use mcp-label-studio?

Install via uv (recommended) by cloning the repository and running uv pip install -e . and uv run server.py, or use Docker with docker build and docker run setting the LABEL_STUDIO_API_KEY and LABEL_STUDIO_URL environment variables. Integrate with Claude Desktop by adding the server configuration (uv or docker variant) to claude_desktop_config.json.

Key features of mcp-label-studio

  • Create, update, delete annotation projects in Label Studio
  • List all projects and get detailed project information
  • Export project annotations (JSON, CSV, TSV, CONLL2003, COCO)
  • Import tasks from a file into a project
  • Retrieve supported export formats for a project

Use cases of mcp-label-studio

  • Automate the creation and maintenance of annotation projects for ML pipelines
  • Batch import labeling tasks from local files into Label Studio
  • Export completed annotations in various formats for downstream model training
  • Integrate labeling workflows into larger MCP-based AI applications

FAQ from mcp-label-studio

How do I install mcp-label-studio?

Use uv by cloning the repository, running uv pip install -e ., then uv run server.py. Alternatively, build a Docker image with docker build -t mcp-label-studio . and run it with the required environment variables.

What environment variables are required?

You must set LABEL_STUDIO_API_KEY (your Label Studio API key) and LABEL_STUDIO_URL (the URL of your Label Studio instance).

What tools does mcp-label-studio provide?

Eight tools: get_projects, get_project_detail, create_project, update_project, delete_project, export_project_annotations, import_tasks_from_file, and get_export_formats.

How can I debug the server?

Use the MCP inspector: LABEL_STUDIO_API_KEY=YOUR_API_KEY LABEL_STUDIO_URL=YOUR_URL npx @modelcontextprotocol/inspector uv run server.py from the project directory.

What license is mcp-label-studio under?

It is licensed under the MIT License, allowing free use, modification, and distribution.

评论

其他 分类下的更多 MCP 服务器

Inbox Zero AI MCP

elie222

The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.

11.6K

Inbox Zero AI

elie222

The world's best AI personal assistant for email. Open source app to help you reach inbox zero fast.

11.6K

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K

🪟 Windows-MCP

CursorTouch

MCP Server for Computer Use in Windows

6.3K

FastMCP v2 🚀

jlowin

🚀 The fast, Pythonic way to build MCP servers and clients.

26.0K