Roboflow MCP Server

Claude Code (CLI)

claude mcp add roboflow \
  --transport http https://mcp.roboflow.com/mcp \
  --header "x-api-key: YOUR_ROBOFLOW_API_KEY" \
  --header "Accept: application/json, text/event-stream"Copy

Claude Desktop / Generic MCP Client

Add to your config (e.g. claude_desktop_config.json):

{
  "mcpServers": {
    "roboflow": {
      "type": "http",
      "url": "https://mcp.roboflow.com/mcp",
      "headers": {
        "x-api-key": "YOUR_ROBOFLOW_API_KEY",
        "Accept": "application/json, text/event-stream"
      }
    }
  }
}Copy

Get your API key at https://app.roboflow.com/settings/api.

What this server provides

30 tools across these categories:

• Projects — manage projects in your workspace
  projects_list, projects_create, projects_get
• Images — prepare image uploads for a project
  images_prepare_upload, images_prepare_upload_zip, images_upload_zip_status, images_search
• Annotations — save annotations to a project image
  annotations_save
• Batch — organize images into batches and create labeling jobs
  annotation_batches_list, annotation_batches_get, annotation_jobs_create
• Versions — create and inspect dataset versions
  versions_generate, versions_get, versions_export
• Models — train models and monitor training progress
  models_list, models_get, models_infer, models_train, models_get_training_status
• Workflows — build and execute inference pipelines
  workflows_list, workflows_get, workflows_create, workflows_update, workflow_blocks_list, workflow_blocks_get_schema, workflow_specs_validate, workflows_run, workflow_specs_run
• Universe — search public datasets on Roboflow Universe
  universe_search
• Meta — report issues or suggestions
  meta_feedback_send

Skills (expert knowledge as MCP resources)

Connected clients can read these as MCP resources for guidance on common tasks:

• api-reference — Reference for Roboflow REST API and Inference API — hosts (api.roboflow.com, serverless.roboflow.com, dedicated, localhost:9001), auth, and request/response formats.
  inference, rest-api
• data-management — Use when uploading images, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification), tags, splits, versions, or RoboQL search.
  labeling
• inference — Use when running Roboflow model inference or choosing deployment (serverless, dedicated, self-hosted, batch); prefer Workflows over direct model calls.
  workflow-templates, workflows
• plans-and-pricing — Use when answering questions about Roboflow plans, credit usage, or cost estimation; directs users to roboflow.com/pricing for current dollar amounts.
• product-navigation — Use when explaining where Roboflow features live in the app.roboflow.com web app, mapping intents like upload, annotate, train, deploy to specific page URLs.
  features-by-page
• training-and-evaluation — Use when training Roboflow models or improving accuracy - covers architecture selection, model IDs, checkpoints, evaluation metrics, and the iterative improvement playbook.
  improvement-playbook
• universe — Use when searching for or using public datasets/models on Roboflow Universe (universe.roboflow.com), the open repository of 1M+ computer vision datasets and 50K+ pre-trained models.

For AI agents

A short, LLM-friendly summary of this server lives at /llms.txt following the llms.txt convention.