Label Studio MCP Server

Example Use Cases

• 1

  Programmatically create and configure Label Studio projects (including uploading XML label configs) to automate setup of labeling tasks.

• 2

  Bulk-import task JSON payloads into a Label Studio project and list or retrieve task data and annotations for downstream processing.

• 3

  Attach model predictions to specific Label Studio tasks (including model versioning and scores) and query task/annotation status.

Description

This MCP server provides tools to create, update, list, and inspect Label Studio projects and tasks, import task payloads from JSON files, and attach model predictions to tasks. It integrates with a running Label Studio instance using an API key and the official label-studio-sdk to automate labeling workflows. The server is useful for agents that need to manage labeling pipelines, bulk-import tasks, and add or query model predictions without direct UI interaction.