The Scale ai MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your Scale ai account. It provides structured and secure access to your projects, batches, annotation tasks, and team information, so your agent can create labeling tasks, organize work, monitor progress, and review project details on your behalf.
- Create image and video annotation tasks: Have your agent set up image, video, video playback, pixel segmentation, LiDAR annotation, or LiDAR segmentation work for your datasets.
- Prepare text and document labeling: Direct your agent to create document transcription, named entity recognition, and text collection tasks for training data.
- Organize annotation workloads: Let the agent create project batches, add tags to existing tasks, and assign unique references to keep work organized.
- Monitor batch progress: Instruct your agent to check batch status, review completion counts, and retrieve batch details.
- Review projects, tasks, and teams: Have your agent list projects, batches, and tasks, inspect their details, and retrieve basic information about team members.