# Roboflow MCP Server

> A hosted Model Context Protocol (MCP) server that exposes Roboflow's computer vision platform — projects, datasets, images, annotations, versions, models, and Workflows — as tools for AI assistants.

This server speaks MCP over streamable HTTP at `https://mcp.roboflow.one/mcp`. Clients can connect via an MCP URL only and will be prompted to sign in to Roboflow.

## Connection
- [MCP endpoint](https://mcp.roboflow.one/mcp): streamable-http transport, supports MCP OAuth discovery
- [OAuth protected-resource metadata](https://mcp.roboflow.one/.well-known/oauth-protected-resource/mcp)
- [Homepage](https://mcp.roboflow.one/): install snippets for Claude Code, Claude Desktop, and other MCP clients

## Apps
- **Apps** — Prefab MCP apps when params are unclear, UX matters, or humans pick from real data
  - `create_project_app`: Use this when project settings are unknown, debatable, or should be chosen or reviewed in a form—not inferred entirely by the agent. MCP alone lacks the UX to settle those fields confidently; the UI collects name, project type, annotation label text, and license before anything is created.
  - `universe_search_app`: Use this when the next step needs human judgment or visible UX: exploring queries, comparing public projects, opening thumbnails/previews, or confirming which dataset to fork into the workspace. Choosing among datasets is data the user must see; ranked JSON from ``universe_search`` alone usually cannot substitute for that decision surface.

## Tool categories
- **Agent** — chat with the Roboflow AI agent for Roboflow Q&A, advanced Workflow building, and CV solution planning
  - `agent_chat`: Chat with the Roboflow AI agent.
  - `agent_conversations_list`: List Roboflow agent conversations in the workspace.
  - `agent_conversation_get`: Get a single agent conversation with its full message
  - `agent_workflow_publish`: Publish the latest agent-edited draft of a workflow.
- **Projects** — manage projects in your workspace
  - `projects_list`: List all projects in the workspace associated with the API key.
  - `projects_create`: Create a new computer vision project.
  - `projects_health`: Get dataset health check for a project.
  - `projects_get`: Get detailed info about a project including versions, classes, splits, and trained models.
  - `projects_fork`: Enqueue an async fork of a public Universe project into your workspace.
- **Images** — upload, search, and manage images across projects
  - `images_prepare_upload`: Get an upload URL to upload a single image to a project.
  - `images_prepare_upload_zip`: Prepare a zip upload of images and annotations to a project.
  - `images_upload_zip_status`: Check the status of a zip upload task.
  - `images_update_metadata`: Update metadata and tags on a single image. Metadata and tags can be useful to find images later in the platform.
  - `images_batch_update_metadata`: Batch-update metadata and tags on multiple images (async).
  - `images_workspace_search`: Search images across the entire workspace using RoboQL.
  - `images_search`: Search for images inside a project.
  - `universe_dataset_images_search`: Search images inside a public Universe dataset URL.
- **Annotations** — save annotations to a project image
  - `annotations_save`: Save an annotation for an existing image.
- **Auto-label** — kick off and track hosted auto-labeling jobs
  - `autolabel_start`: Start a hosted auto-label job over a batch of images.
  - `autolabel_job_get`: Get per-subjob status and progress for an auto-label job.
- **Batch** — organize images into batches and create labeling jobs
  - `annotation_batches_list`: List upload batches in a project.
  - `annotation_batches_get`: Get details about a specific batch.
  - `annotation_jobs_create`: Create an annotation job to assign a batch of images to a labeler.
- **Versions** — create and inspect dataset versions
  - `versions_generate`: Create a new dataset version with optional preprocessing and augmentation.
  - `versions_get`: Get info about a dataset version including splits and model metrics.
  - `versions_export`: Check or trigger a dataset export for a version.
- **Models** — train models and monitor training progress
  - `models_list`: List trained models associated with a project.
  - `models_get`: Get details for a trained model.
  - `models_infer`: Run hosted inference on an image using a trained model.
  - `models_train`: Start training a model on a dataset version.
  - `models_get_training_status`: Get the training progress and metrics for a dataset version.
  - `trainings_get_results`: Run-level training results bundle.
  - `trainings_cancel`: Cancel an in-flight training run.
  - `trainings_stop`: Request an early stop on an in-flight training run.
  - `models_star_nas`: Star or unstar a NAS-trained model.
- **Model Evaluations** — inspect held-out evaluation results for trained models — mAP, precision/recall sweeps, per-class breakdowns, confusion matrices, vector clustering, and per-image predictions
  - `model_evals_list`: List model evaluations in the current workspace.
  - `model_evals_get`: Get the top-level summary for a single model evaluation.
  - `model_evals_get_map_results`: Get per-split mAP results for an eval.
  - `model_evals_get_confidence_sweep`: Get the precision/recall/F1 confidence sweep for an eval.
  - `model_evals_get_performance_by_class`: Get per-class performance metrics for a single split.
  - `model_evals_get_confusion_matrix`: Get the confusion matrix for an eval.
  - `model_evals_get_vector_analysis`: Get UMAP + HDBSCAN clustering of image embeddings for an eval.
  - `model_evals_get_image_predictions`: Get per-image prediction stats for an eval (paginated).
  - `model_evals_get_recommendations`: Get the LLM-generated recommendations for an eval, if available.
- **Workflows** — build and execute inference pipelines
  - `workflows_list`: List saved workflows in the current workspace.
  - `workflows_get`: Get details for a saved workflow.
  - `workflows_create`: Create and save a new Workflow in the workspace.
  - `workflows_update`: Update an existing saved Workflow's name and definition.
  - `workflow_blocks_list`: List all available Workflow blocks with a short summary of each.
  - `workflow_blocks_get_schema`: Get the full schema of a specific Workflow block.
  - `workflow_specs_validate`: Validate a Workflow JSON definition without executing it.
  - `workflows_run`: Execute a saved Workflow on one or more images.
  - `workflow_specs_run`: Execute a Workflow from an inline JSON definition.
- **Devices** — list, inspect, and provision Roboflow Deployment Manager devices via the external Deployments v2 API
  - `devices_list`: List devices registered in the workspace.
  - `devices_create`: Provision a new v2 device.
  - `devices_get`: Get a single device by id.
  - `devices_get_config`: Get the device's current runtime configuration.
  - `devices_get_default_config`: Get the workspace's default device configuration.
  - `devices_update_config`: Update the device's runtime configuration.
  - `devices_get_config_history`: List prior configuration revisions, newest first.
  - `devices_streams_list`: List streams configured on the device.
  - `devices_streams_get`: Get a single stream by id.
  - `devices_get_logs`: Fetch device logs from Elasticsearch.
  - `devices_get_telemetry`: Get aggregated hardware metrics (cpu, memory, disk, gpu).
  - `devices_get_events`: List device and stream lifecycle events.
- **Universe** — search public datasets on Roboflow Universe
  - `universe_search`: Search Roboflow Universe for datasets or models.
- **Async tasks** — poll the status of long-running async operations
  - `async_tasks_get`: Poll an async task by id. Poll every 5 seconds; processing may take up to 30 seconds to start.
- **Meta** — report issues or suggestions
  - `meta_feedback_send`: Report a bug, missing feature, UX friction, or documentation issue.

## Skills (expert knowledge as MCP resources)
- **api-reference** — Protocol-level facts for Roboflow REST and Inference APIs — URL patterns, auth, parameters, error codes, and SDK quick-start. For deployment strategy and Workflow execution patterns, see roboflow-inference.
  - [api-reference/SKILL](roboflow://skills/api-reference/SKILL)
  - [api-reference/inference](roboflow://skills/api-reference/inference)
  - [api-reference/rest-api](roboflow://skills/api-reference/rest-api)
- **data-management** — Use when uploading images, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification), tags, splits, versions, or RoboQL search.
  - [data-management/SKILL](roboflow://skills/data-management/SKILL)
  - [data-management/labeling](roboflow://skills/data-management/labeling)
- **inference** — Deployment option comparison (serverless, dedicated, self-hosted, batch) and Workflow execution patterns. For raw API URL patterns, auth, and request/response formats, see roboflow-api-reference.
  - [inference/SKILL](roboflow://skills/inference/SKILL)
  - [inference/batch-jobs](roboflow://skills/inference/batch-jobs)
  - [inference/batch-staging](roboflow://skills/inference/batch-staging)
  - [inference/local-tooling](roboflow://skills/inference/local-tooling)
  - [inference/workflow-templates](roboflow://skills/inference/workflow-templates)
  - [inference/workflows](roboflow://skills/inference/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.
  - [plans-and-pricing/SKILL](roboflow://skills/plans-and-pricing/SKILL)
- **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.
  - [product-navigation/SKILL](roboflow://skills/product-navigation/SKILL)
  - [product-navigation/features-by-page](roboflow://skills/product-navigation/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.
  - [training-and-evaluation/SKILL](roboflow://skills/training-and-evaluation/SKILL)
  - [training-and-evaluation/improvement-playbook](roboflow://skills/training-and-evaluation/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.
  - [universe/SKILL](roboflow://skills/universe/SKILL)

## Optional
- [Roboflow platform documentation](https://docs.roboflow.com)
- [Model Context Protocol specification](https://modelcontextprotocol.io)
- [Roboflow homepage](https://roboflow.com)