# 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.com/mcp`. Clients can connect via an MCP URL only and will be prompted to sign in to Roboflow.

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

## Apps
- **Apps** — Interactive Roboflow apps.
  - `annotation_editor_open`: Use this when a person needs to inspect or edit image annotations visually. Changes are saved only when the person chooses Save in the widget.
  - `annotation_editor_read`: Reload the editor snapshot and revision after a save or conflict.
  - `annotation_editor_save`: Save the user's annotation edits using the loaded group revision.

## 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_chat_result`: Collect the result of an agent_chat run that was still working.
  - `agent_conversations_list`: List Roboflow agent conversations in the workspace.
  - `agent_conversation_get`: Get a single agent conversation with its message history.
  - `agent_workflow_publish`: Publish the latest agent-edited draft of a workflow.
- **Projects** — manage projects in your workspace
  - `projects_list`: List projects in the workspace associated with the credential.
  - `projects_create`: Create a new computer vision project, including Action Recognition.
  - `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_delete`: Delete a project, moving it to the workspace Trash.
  - `projects_fork`: Enqueue an async fork of a public Universe project into your workspace.
  - `datasets_rebalance_splits`: Enqueue an async rebalance of a project's train/valid/test splits.
- **Project Deployment** — launch and manage a stable live endpoint for a project
  - `project_deployment_get`: Get the stable live endpoint state for a project.
  - `project_deployment_launch`: Create or prepare a stable Project Deployment.
  - `project_deployment_set_model`: Change the default model behind a Project Deployment without changing its endpoint.
  - `project_deployment_run`: Run inference through the project's stable live endpoint.
  - `project_deployment_enable_active_learning`: Enable Active Learning for a Project Deployment.
  - `project_deployment_disable_active_learning`: Pause Active Learning collection for a Project Deployment.
  - `project_deployment_configure_active_learning`: Configure how a Project Deployment collects Active Learning data.
  - `project_deployment_list_review_queues`: List Active Learning review queues fed by production inference.
- **Images** — upload images and videos, search, and manage images across projects
  - `image_upload`: Prepare a signed URL for uploading a zip of images and optional annotations.
  - `image_upload_status`: Check the status of an image zip upload task.
  - `video_upload`: Prepare a signed URL for uploading one local video (MP4 or MOV) into a project.
  - `video_upload_status`: Check the ingestion status of a video uploaded with video_upload.
  - `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 images in one project by what they show, with optional filters.
  - `universe_dataset_images_search`: Search images inside a public Universe dataset given its URL.
- **Media Understanding** — analyze images and short videos with a vision model to scope CV tasks, suggest classes, and plan workflows
  - `media_analyze`: Analyze an image or short video with a vision model to understand its contents.
  - `media_upload`: Prepare a local image or video for analysis with media_analyze.
  - `media_trim`: Cut a clip from an Asset Library video and get an analyzable public url.
  - `media_upload_finalize`: Validate a staged upload (from media_upload) and publish it.
- **Annotations** — save image annotations and Action Recognition video segments
  - `video_segments_annotate`: Annotate one Action Recognition video Source with a complete video-coco document.
  - `annotations_save`: Save an annotation for an existing image.
- **Auto-label** — kick off and track hosted auto-labeling jobs
  - `autolabel_models_list`: List the foundation models available for hosted auto-labeling in this workspace.
  - `autolabel_preview`: Preview one image with a foundation model to judge it before starting an auto-label job.
  - `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.
- **Annotation Administration** — explicitly manage annotation batches, assignments, review, and Dataset acceptance
  - `annotation_board_status`: Summarize a project's Annotate board: Unassigned, Annotating, Review, and Dataset.
  - `annotation_batches_list`: List upload batches in a project.
  - `annotation_batches_get`: Get details about one upload batch.
  - `annotation_batches_admin_list`: List annotation-board batches with cursor pagination.
  - `annotation_batches_admin_get`: Get details about one annotation-board batch.
  - `annotation_batches_admin_images_list`: List image IDs in a batch with cursor pagination.
  - `annotation_batches_create`: Move selected images from one batch into a new batch.
  - `annotation_batches_merge`: Move all images from source batches into a target batch and remove the emptied sources.
  - `annotation_batches_delete`: Delete a batch. By default its images move to unassigned and remain accessible.
  - `annotation_jobs_list`: List annotation jobs and return a continuation token when more exist.
  - `annotation_jobs_get`: Get details about one annotation job.
  - `annotation_jobs_images_list`: List image IDs assigned to a job with cursor pagination.
  - `annotation_jobs_create`: Create a job and move images from a batch into it.
  - `annotation_jobs_images_reassign`: Create a job from selected images, removing their prior assignment or Dataset membership.
  - `annotation_jobs_images_add`: Move images into an existing job and reconcile assignment and Dataset counters.
  - `annotation_jobs_update`: Update exactly one of a job's labeler, reviewer, or instructions.
  - `annotation_jobs_submit_for_review`: Advance a labeling job into review.
  - `annotation_jobs_return_for_edits`: Move a review job back to labeling, optionally with a new labeler.
  - `annotation_jobs_review_image`: Set the explicit review status for one image in a job.
  - `annotation_jobs_review_images`: Set a status for every job image matching the supplied current status.
  - `annotation_jobs_accept_into_dataset`: Finalize selected job images into the Dataset and assign splits.
  - `annotation_jobs_move_to_unassigned`: Remove a job while moving its images to an unassigned batch so they remain accessible.
  - `annotation_jobs_delete_annotations`: Delete the project's annotation data for every image assigned to a job.
- **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 its trainings.
  - `versions_delete`: Delete a dataset version, moving it to the workspace Trash.
  - `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.
  - `trainings_describe_recipe`: Describe the tuning surface for a model type and get a ready-to-submit train recipe.
  - `trainings_create`: Create a training on a dataset version.
  - `models_custom_weights_upload_guide`: Get the recipe for uploading locally trained model weights to Roboflow.
  - `trainings_list`: List the trainings on a dataset version.
  - `trainings_get`: Get a training's status, produced models, and metrics.
  - `trainings_cancel`: Cancel an in-flight training run.
  - `trainings_stop`: Request an early stop on an in-flight training run.
  - `trainings_delete`: Delete a training run, moving it to the workspace Trash.
  - `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, per-image predictions, and accuracy/latency comparisons
  - `model_evals_compare`: Compare models trained on or uploaded to one dataset version.
  - `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.
  - `workflows_delete`: Delete a saved Workflow, moving it to the workspace Trash.
  - `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.
- **Workflow Evals** — design evaluations for Workflows: discover evaluators and schemas, author Specs and Eval Datasets, manage Cases, and confirm Workflow output bindings
  - `workflow_evals_capabilities`: Describe the Workflow Evals API: operations, schemas, evaluators, and caller actions.
  - `workflow_evals_evaluators_list`: List the evaluators available to Spec Checks in the installed engine.
  - `workflow_evals_evaluators_get`: Get one evaluator's description and configuration JSON Schema.
  - `workflow_evals_schema_get`: Get a JSON Schema for authoring Workflow Evals payloads.
  - `workflow_evals_openapi_get`: Get the machine-readable OpenAPI 3.1 contract for the Workflow Evals API.
  - `workflow_evals_agent_manifest`: Get the engine's agent manifest. Start here before designing or drafting an Eval.
  - `workflow_evals_agent_resources_list`: List engine-owned agent skills or bundles. Returns ``{engineVersion, items}``.
  - `workflow_evals_agent_resource_get`: Get one engine-owned agent resource.
  - `workflow_evals_operation_get`: Poll a Workflow Evals deletion operation.
  - `workflow_evals_list`: List Evals in the workspace, newest first.
  - `workflow_evals_create`: Create an Eval. Every field is optional; an empty call creates a draft.
  - `workflow_evals_get`: Get one Eval by ID or by its readable URL (exactly one).
  - `workflow_evals_update`: Update an Eval under optimistic concurrency.
  - `workflow_evals_delete`: Delete an Eval with all its Runs, Executions, results, and owned artifacts.
  - `workflow_evals_specs_list`: List live Specs in the workspace. Returns ``{items, nextCursor?}``.
  - `workflow_evals_specs_create`: Validate, compile, and save a Spec. Invalid Specs return 422 with engine diagnostics.
  - `workflow_evals_specs_get`: Get the current live Spec and its ``revision``.
  - `workflow_evals_specs_update`: Replace a live Spec's policy and/or metadata. The full result is recompiled.
  - `workflow_evals_specs_validate`: Validate an unsaved Spec: compile diagnostics plus derived ground-truth requirements.
  - `workflow_evals_specs_compatible_datasets`: List Eval Datasets whose inputs and ground truth satisfy a Spec.
  - `workflow_evals_datasets_list`: List Eval Datasets with declared contracts, Case counts, slices, and completeness.
  - `workflow_evals_datasets_create`: Create an empty Eval Dataset with declared input and ground-truth contracts.
  - `workflow_evals_datasets_get`: Get an Eval Dataset and its derived summary (no Cases inline) plus ``revision``.
  - `workflow_evals_datasets_update`: Update Eval Dataset metadata (name/description).
  - `workflow_evals_datasets_update_contract`: Replace a Dataset's declared input and/or ground-truth contract.
  - `workflow_evals_datasets_compatible_specs`: List Specs whose input and ground-truth requirements this Dataset satisfies.
  - `workflow_evals_cases_list`: List Cases in an Eval Dataset (summaries, no media bytes).
  - `workflow_evals_cases_create`: Add one Case to an Eval Dataset, validated against the Dataset's contracts.
  - `workflow_evals_cases_get`: Get one complete Case with short-lived read URLs for its assets and its ``revision``.
  - `workflow_evals_cases_replace`: Replace a Case atomically (full replacement, not a field patch).
  - `workflow_evals_cases_delete`: Remove a Case from the live Dataset. Runs that already froze it stay readable.
  - `workflow_evals_cases_import`: Bulk-import images from Roboflow sources, project datasets, or uploaded artifacts.
  - `workflow_evals_case_imports_get`: Poll a Case import.
  - `workflow_evals_case_assets_upload`: Start uploading a local image as Case input. The MCP server never receives bytes.
  - `workflow_evals_case_assets_upload_complete`: Verify and publish an uploaded Case image. Returns the stable ``artifactId``.
  - `workflow_evals_bindings_suggest`: Propose a BindingSet mapping a Spec's inputs/outputs onto a Workflow Subject.
  - `workflow_evals_bindings_validate`: Validate a BindingSet against a Spec, Dataset, and Subject.
- **Workflow Eval Runs** — run Workflow Evals against one or more Workflows, prepare candidate outputs, inspect and compare per-Case results, export them, and analyze image embeddings
  - `workflow_evals_case_preparations_start`: Run the Eval's Workflow over Dataset Cases to produce candidate outputs. Costs inference.
  - `workflow_evals_case_preparations_get`: Get a Case preparation's state and progress (current one when no ID is given).
  - `workflow_evals_case_preparations_resume`: Resume a preparation, keeping successful Cases and requeueing only failed ones.
  - `workflow_evals_case_preparations_cases`: List candidate outputs per Case. Each success has a short-lived ``candidateOutput.readUrl``.
  - `workflow_evals_runs_list`: List an Eval's Runs, newest first. ``limit=1`` gives the latest Run.
  - `workflow_evals_runs_start`: Start one grouped Run that evaluates every Subject on the Eval's Spec and Dataset.
  - `workflow_evals_runs_get`: Get a Run's lifecycle state, child Execution counts, and ``retry`` eligibility.
  - `workflow_evals_runs_configuration`: Get a Run's frozen shared configuration: Spec, Dataset snapshot, and Case selection.
  - `workflow_evals_runs_cancel`: Request cancellation of a queued or running Run. Idempotent.
  - `workflow_evals_runs_retry_failed`: Retry only missing or operationally failed Cases in the same Run.
  - `workflow_evals_runs_replay`: Create a new derived Run from a source Run, optionally with new Subjects.
  - `workflow_evals_runs_rescore`: Score retained Workflow outputs against a revised Spec without re-running Workflows.
  - `workflow_evals_runs_delete`: Delete a Run with its Executions and results.
  - `workflow_evals_executions_list`: List a Run's Executions (one per evaluated Workflow Subject).
  - `workflow_evals_executions_get`: Get one Execution's lifecycle, frozen Subject identity, and result availability.
  - `workflow_evals_executions_configuration`: Get an Execution's frozen Subject manifest, BindingSet, and linked plan.
  - `workflow_evals_executions_input_cases`: Page through an Execution's input Cases for a client-owned (local) runner.
  - `workflow_evals_executions_submit_capture`: Submit one Case's Workflow output from a local runner. The platform scores it.
  - `workflow_evals_results_overview`: Get an Execution's aggregate per-Check performance and runtime summary.
  - `workflow_evals_results_cases_list`: List per-Case results for an Execution with compact ``checkResults`` by Check ID.
  - `workflow_evals_results_case_get`: Get one Case's result: inputs, frozen ground truth, bound actual values, and Check results.
  - `workflow_evals_results_compare`: Compare completed Executions: per-Check deltas and a page of Case-level deltas.
  - `workflow_evals_exports_start`: Start a server-side export of a Run's results.
  - `workflow_evals_exports_get_job`: Poll an export. ``{asyncTaskId, state, progress, result?}``; ``result`` has the artifact and download URL.
  - `workflow_evals_exports_download_url`: Get a fresh short-lived download URL for a completed export.
  - `workflow_evals_embeddings_start`: Generate (or reuse) a 2D embedding projection of an Execution's input images.
  - `workflow_evals_embeddings_get`: Get embedding analysis jobs. Never launches analysis.
  - `workflow_evals_embeddings_cancel`: Request cooperative cancellation of an embedding job.
- **Batch Processing** — run Workflows over huge image/video sets: Asset Library selections (platform-staged) or batches you stage yourself
  - `batch_processing_asset_library_job_create`: Queue a durable Workflow run over an Asset Library selection.
  - `batch_processing_asset_library_task_get`: Poll the launch receipt until its durable job is registered.
  - `batch_processing_guide`: Route a bulk image/video processing request onto batch processing: pick the staging mode, settle the batch id, and return the exact inference-cli commands for the whole run.
  - `batch_processing_staging_batches_list`: List the workspace's data-staging batches (inputs and job results).
  - `batch_processing_staging_batch_get`: Get a staged batch's metadata, file count, and ingest progress.
  - `batch_processing_staging_batch_files_list`: List a staged batch's files with time-limited signed download URLs.
  - `batch_processing_jobs_list`: List the workspace's batch processing jobs.
  - `batch_processing_job_get`: Get a job's status and per-stage progress.
  - `batch_processing_staged_job_start`: Start a batch processing job: run a Workflow over a staged batch.
  - `batch_processing_job_abort`: Abort a running batch processing job. Already-produced partial
  - `batch_processing_job_restart`: Restart a failed or aborted job, optionally overriding compute
  - `batch_processing_job_logs`: Fetch a job's processing logs (paginated). Use this to diagnose
  - `batch_processing_run`: Master tool: drive a batch processing run over a staging batch you name (stage -> ingest -> job -> results) through repeated resumable calls. For Asset Library selections, use batch_processing_asset_library_job_create instead.
- **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 and device health sections.
  - `devices_get_deployments`: Get the device's deployment history with per-deployment activity.
  - `devices_get_exception_occurrences`: List raw occurrences of one exception fingerprint, newest first.
  - `devices_get_snapshot`: Get the full current state of a device in one call.
  - `devices_get_events`: List device and stream lifecycle events.
  - `devices_get_services`: List the sidecar containers a device runs and what each is doing.
  - `devices_get_plc_status`: Answer "is this PLC connected?" for a device's PLC relay sidecars.
  - `devices_get_event_store_stats`: Report a device's local event store against its capacity limits.
  - `devices_get_service_series`: Chart one named sidecar metric over time.
- **API Keys** — manage workspace API keys
  - `api_keys_list`: List all API keys for the workspace.
  - `api_keys_get`: Get metadata for a single API key by its keyId.
  - `api_keys_get_publishable`: Get the workspace's publishable key (``rf_<workspaceId>``).
  - `api_keys_create`: Create a new API key for the workspace.
  - `api_keys_update`: Update the name, scopes, or custom_metadata of an existing API key.
  - `api_keys_protect`: Mark an API key as protected so it cannot be revoked or disabled.
  - `api_keys_disable`: Disable (or re-enable) an API key without revoking it.
  - `api_keys_revoke`: Permanently revoke (delete) an API key.
  - `api_keys_roll`: Roll (rotate) an API key: revoke it and mint a replacement in one step.
- **Folders** — list the workspace folders visible to the credential
  - `folders_list`: List the workspace folders visible to this credential.
- **Vision Events** — query and report on production vision events (quality checks, inventory counts, safety alerts, operator feedback) and manage the use cases they are grouped under and the filters saved for them
  - `vision_events_query`: Query production vision events for a use case, with filters and pagination.
  - `vision_events_use_cases_list`: List the vision-event use cases in the workspace.
  - `vision_events_custom_metadata_schema_get`: Get the custom-metadata schema discovered for a use case.
  - `vision_events_image_metadata_schema_get`: Get the per-image metadata schema discovered for a use case.
  - `vision_events_use_case_create`: Create a new vision-event use case.
  - `vision_events_use_case_rename`: Rename an existing vision-event use case.
  - `vision_events_use_case_archive`: Archive a vision-event use case.
  - `vision_events_use_case_unarchive`: Restore a previously archived vision-event use case.
  - `vision_events_saved_filters_list`: List the filters saved for a use case.
  - `vision_events_saved_filter_get`: Get one saved filter. Returns ``{savedFilter}``.
  - `vision_events_saved_filter_create`: Save a named filter for a use case, shared with the whole workspace.
  - `vision_events_saved_filter_update`: Rename a saved filter, save over its filters, or both.
  - `vision_events_saved_filter_delete`: Delete a saved filter for everyone in the workspace.
- **Universe** — search public datasets on Roboflow Universe
  - `universe_search`: Search Roboflow Universe for datasets or models.
- **Datasources** — configure and run cloud-bucket mirroring into projects
  - `datasources_list`: List datasource (cloud-bucket mirror) configurations in the workspace.
  - `datasource_get`: Get full detail for a single datasource, including its bucket,
  - `datasource_validate`: Check that Roboflow can access the datasource's bucket.
  - `datasource_job_get`: Get the status and statistics of a single datasource mirror run.
  - `datasource_create`: Create a datasource that mirrors a cloud bucket into your workspace.
  - `datasource_update`: Update a datasource. Only the fields you pass are changed.
  - `datasource_delete`: Delete a datasource configuration. Does not delete already-mirrored images.
  - `datasource_trigger`: Start a mirror run for a datasource.
  - `connect_cloud_storage`: Connect a cloud bucket end to end in one call: reuse or create a credential, create the datasource, validate it, and start the first mirror run. A run can overwrite imported images or metadata and remove mirrored images whose source disappeared.
- **Credentials** — manage cloud-storage credentials used by datasources
  - `credentials_list`: List cloud-storage credentials in the workspace.
  - `credentials_delete`: Delete a credential. Datasources still referencing it will fail to run.
  - `get_aws_storage_credential_setup_link`: Get a Roboflow app link to set up an AWS S3 storage credential.
- **Trash** — list and restore soft-deleted items
  - `trash_list`: List soft-deleted items currently in the workspace Trash.
  - `trash_restore_project`: Restore a soft-deleted project from the workspace Trash.
  - `trash_restore_version`: Restore a soft-deleted dataset version from the workspace Trash.
  - `trash_restore_workflow`: Restore a soft-deleted workflow from the workspace Trash.
  - `trash_restore_training`: Restore a soft-deleted training run from the workspace Trash.
- **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`: Send product feedback about Roboflow to the Roboflow team.

## Skills (expert knowledge as MCP resources)
- **draw-vision** — Draw computer vision annotations, labels, and predictions over images and videos using supervision-js in JavaScript/TypeScript or Supervision in Python.
  - [draw-vision/SKILL](roboflow://skills/draw-vision/SKILL)
- **roboflow-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.
  - [roboflow-api-reference/SKILL](roboflow://skills/roboflow-api-reference/SKILL)
  - [roboflow-api-reference/api-key-management](roboflow://skills/roboflow-api-reference/api-key-management)
  - [roboflow-api-reference/inference](roboflow://skills/roboflow-api-reference/inference)
  - [roboflow-api-reference/rest-api](roboflow://skills/roboflow-api-reference/rest-api)
- **roboflow-batch-processing** — Use when running a Roboflow Workflow over many images or videos with batch processing — Asset Library jobs (the platform stages workspace images for you) or staged jobs (you stage external local/cloud files with the inference-cli), plus monitoring jobs, downloading results, and routing imports to datasources (bucket mirror) instead.
  - [roboflow-batch-processing/SKILL](roboflow://skills/roboflow-batch-processing/SKILL)
- **roboflow-cloud-storage** — Use when connecting cloud storage (AWS S3 / S3-compatible or Google Cloud Storage) to Roboflow to mirror images into a workspace — creating storage credentials, defining datasources (bucket-mirror configs), selecting objects with glob rules, validating access, and running/scheduling mirror jobs.
  - [roboflow-cloud-storage/SKILL](roboflow://skills/roboflow-cloud-storage/SKILL)
- **roboflow-custom-weights-upload** — Use when uploading locally trained model weights (YOLO, RF-DETR, YOLO-NAS, PaliGemma, Florence-2) to Roboflow — the client-side Python SDK upload flow, safe API key handling, per-family packaging requirements, and verifying the upload.
  - [roboflow-custom-weights-upload/SKILL](roboflow://skills/roboflow-custom-weights-upload/SKILL)
- **roboflow-data-management** — Use when uploading images, labeling, organizing datasets, creating Roboflow projects (detection/segmentation/keypoint/classification), tags, splits, versions, or RoboQL search.
  - [roboflow-data-management/SKILL](roboflow://skills/roboflow-data-management/SKILL)
  - [roboflow-data-management/labeling](roboflow://skills/roboflow-data-management/labeling)
- **roboflow-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.
  - [roboflow-inference/SKILL](roboflow://skills/roboflow-inference/SKILL)
  - [roboflow-inference/batch-jobs](roboflow://skills/roboflow-inference/batch-jobs)
  - [roboflow-inference/batch-staging](roboflow://skills/roboflow-inference/batch-staging)
  - [roboflow-inference/local-tooling](roboflow://skills/roboflow-inference/local-tooling)
  - [roboflow-inference/workflow-templates](roboflow://skills/roboflow-inference/workflow-templates)
  - [roboflow-inference/workflows](roboflow://skills/roboflow-inference/workflows)
- **roboflow-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.
  - [roboflow-plans-and-pricing/SKILL](roboflow://skills/roboflow-plans-and-pricing/SKILL)
- **roboflow-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.
  - [roboflow-product-navigation/SKILL](roboflow://skills/roboflow-product-navigation/SKILL)
  - [roboflow-product-navigation/features-by-page](roboflow://skills/roboflow-product-navigation/features-by-page)
- **roboflow-training-and-evaluation** — Use when training Roboflow models, diagnosing why a model underperforms, improving accuracy, or setting up a production feedback loop — covers architecture selection, model IDs, checkpoints, Model Evaluation (confusion matrix, per-class metrics, MCP eval tools), the diagnosis-first improvement playbook, and Active Learning through the Project Model Workflow block.
  - [roboflow-training-and-evaluation/SKILL](roboflow://skills/roboflow-training-and-evaluation/SKILL)
  - [roboflow-training-and-evaluation/active-learning](roboflow://skills/roboflow-training-and-evaluation/active-learning)
  - [roboflow-training-and-evaluation/improvement-playbook](roboflow://skills/roboflow-training-and-evaluation/improvement-playbook)
  - [roboflow-training-and-evaluation/model-diagnosis](roboflow://skills/roboflow-training-and-evaluation/model-diagnosis)
  - [roboflow-training-and-evaluation/model-ids](roboflow://skills/roboflow-training-and-evaluation/model-ids)
  - [roboflow-training-and-evaluation/nas-results](roboflow://skills/roboflow-training-and-evaluation/nas-results)
- **roboflow-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.
  - [roboflow-universe/SKILL](roboflow://skills/roboflow-universe/SKILL)
- **workflow-from-eval** — Build or improve a Roboflow Workflow using eval-driven search. Prefer this for open-ended business problems where the approach is unclear, or when comparing alternatives. A business goal plus images or video is enough to start; no existing Workflow or Eval is required. Build an example suite and scorecard the user approves, compare candidate Workflows, then tune and validate the winner. Use direct tools for straightforward builds and edits, Workflow Evals tools for measuring one existing Workflow once, and training skills for training a model.
  - [workflow-from-eval/SKILL](roboflow://skills/workflow-from-eval/SKILL)
  - [workflow-from-eval/gotchas](roboflow://skills/workflow-from-eval/gotchas)
  - [workflow-from-eval/ground-truth](roboflow://skills/workflow-from-eval/ground-truth)
  - [workflow-from-eval/intake](roboflow://skills/workflow-from-eval/intake)
  - [workflow-from-eval/search](roboflow://skills/workflow-from-eval/search)

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