Roboflow

Roboflow MCP Connector for Claude

A+

Manage computer vision workflows — upload images, train models, and manage datasets directly from your AI agent.

29 tools Official Updated Oct 1, 2026 Official Vinkius Partner

Connect Roboflow to your AI agent to streamline your computer vision pipeline. From dataset management to model training and inference, handle your entire CV lifecycle through natural language.

What you can do

  • Workspace & Project Management — List projects, create new ones, or fork from Roboflow Universe to jumpstart your development.
  • Dataset Operations — Upload images (via URL or Base64), manage versions, and download datasets in various formats like COCO or YOLO.
  • Model Training — Start training runs, monitor results, and retrieve precise performance metrics (mAP, precision, recall) for any version.
  • Image Search — Search and filter images within your workspace to audit your data and improve model accuracy.
  • Inference & Results — Run inference on images and retrieve results to verify model behavior in real-time.

How it works

  1. Subscribe to this server
  2. Enter your Roboflow Private API Key
  3. Start building and managing vision models from Claude, Cursor, or any MCP-compatible client

Who is this for?

  • ML Engineers — monitor training progress and dataset health without leaving the terminal or IDE.
  • Data Scientists — quickly query dataset versions and export data for custom training scripts.
  • Product Teams — audit model performance and visualize inference results through simple conversation.
computer-visiondataset-managementmodel-trainingimage-annotationmachine-learningworkflow-automation

29 tools expose this connector's capabilities to your AI agent.

search_workspace_images

Search and filter images within a workspace

get_dataset_health

Requires both workspace and project identifiers. Check dataset health (class distribution, missing annotations, etc)

get_image

Supply the workspace, project, and unique image ID. Get details for a specific image

get_root

Verify authentication and retrieve default workspace

get_training_results

Supply the workspace, project, and version details. Retrieve metrics and status for a version training run

restore_trash

Provide the workspace and the item ID. Restore an item from the trash

run_inference

Provide the dataset ID and version ID. Run inference on an image using hosted models

search_project_images

Search and filter images within a specific project

add_projects_to_folder

Provide the workspace, folder ID, and list of project IDs. Add projects to a folder (Enterprise)

auto_label

Specify the workspace, project, and batch ID. Start an auto-labeling job using foundation models

cancel_training

Provide the workspace, project, and version ID. Cancel an active training job

get_project

Supply both the workspace and project identifiers. Get project details, metadata, and versions

create_annotation_job

Provide the workspace, project, and batch ID. Assign a batch of images to a labeler and reviewer

create_folder

Provide the workspace and desired folder name. Create a project folder (Enterprise)

create_project

Specify the workspace ID and project name. Create a new project in a workspace

delete_images

Specify the workspace, project, and array of image IDs. Delete multiple images from a project

delete_project

Provide the workspace, project, and version ID. Delete a project or version (moves to Trash)

download_dataset

Specify the workspace, project, version, and desired format. Retrieve a download link for a zipped dataset in a specific format

fork_universe_project

Supply the destination workspace and the source URL. Fork a public project from Roboflow Universe

get_async_task

Provide the workspace and task ID. Track long-running operations like forking or large exports

get_version

Provide the workspace, project, and version identifier. Retrieve metadata for a specific dataset version

list_folders

Provide the workspace ID or slug. List project folders in a workspace (Enterprise)

list_trash

Provide the workspace identifier as input. List items in the workspace trash

list_workspace_projects

Provide the workspace ID or slug. List information about a workspace and its projects

manage_image_tags

Provide the workspace, project, image ID, and list of tags. Add, remove, or set tags on an image

start_training

Provide the workspace, project, and version ID. Start training a model on a dataset version

stop_training

Provide the workspace, project, and version ID. Early stop an active training job

upload_annotation

Supply the workspace, project, image ID, and annotation path. Attach an annotation file to an existing image

upload_image

Provide the workspace, project, and image data. Use batch_name for batch identification. Upload an image to a project

See how to talk to your AI agent using Roboflow.

List all projects in my Roboflow workspace 'industrial-safety'.

I've retrieved the projects for 'industrial-safety'. You have 3 active projects: 'Hard Hat Detection' (object-detection), 'Glove Compliance' (classification), and 'Forklift Tracking'. Which one would you like to inspect?

Upload this image URL to the 'Hard Hat Detection' project in workspace 'industrial-safety'.

Uploading image... Success! The image has been added to the 'Hard Hat Detection' project. It is currently in the unassigned batch. Would you like to assign it to a specific split?

Show me the training metrics for version 5 of the 'Forklift Tracking' project.

Fetching results for version 5... The model achieved a mAP of 88.5%, with a precision of 91.2% and recall of 84.7%. The training run is completed. Would you like to see the full metrics breakdown?

You can use the `get_root` tool. It will attempt to authenticate with your key and return the default workspace details if successful.

Related Connectors