Roboflow integration: operations for AI agents
Roboflow connects to Type.com as an integration. This page lists the 178 operations documented for Claude and Codex in Type.com: 86 read, 78 write, 14 destructive.
About Roboflow
Build and use computer vision models fast with Roboflow. Train, deploy, and run inference on object detection, classification, and segmentation models.
Categories: Developer Tools & IT, Design & Media
How to connect Roboflow to Type.com
- Access
- Organization connection: one shared account for the workspace.
To add the connection, follow the steps in Space connections.
Documented operations
178 operations. Your connection's permissions determine which operations are available. API references describe the vendor surface; they do not expand your connection's credentials or access.
178 operations listed.
Cloud Streaming
- Read
List Cloud Streaming sources
Lists registered sources without stream credentials. Temporary sessions are excluded unless requested. The response contains sources only, not jobs or a webcam-availability flag. Requires video-inference-job:read.
- Read
Inspect a Workflow interface
Inspect a saved Workflow before selecting imageOutput and workflowParameters. This does not start processing. Requires video-inference-job:read. Use a workspace API key or OAuth access token for the workspace in the path. Device-bound keys are not accepted.
- Read
Preview raw source video
Requests source demand and placement, not Workflow inference. For connector sources, renew the five-minute demand lease while viewing. Direct-source grants include expiresAt; reacquire before it expires. Returns raw-source playback credentials or an uploaded-file playback URL.
- Read
List Workflow jobs
Lists workspace job records. This route does not expose source, processor or execution credentials. It does not accept source or state filters. Requires video-inference-job:read. Use a workspace API key or OAuth access token for the workspace in the path.
- Read
Get a Workflow job
Read the current job without changing its state. Follow executionGeneration and successorJobId when observing replacement executions. A persistent waiting or blocked job still represents enabled intent until explicitly cancelled. Requires video-inference-job:read.
- Read
Download completed file results
Returns signed URLs after a batch job's result files have been uploaded. This is not live JSON polling, a stream recording endpoint or durable live result history. A JSON-only job may omit video.mp4. URLs expire after one hour. Requires video-inference-job:read.
- Write
Allocate a temporary source
Allocates one WHIP publishing session. The publishing grant admits one publisher within two minutes. At most four unexpired publishing grants may be outstanding per workspace. Repeating the same request and Idempotency-Key returns the same live session, not a renewed grant.
- Write
End a temporary source session
Invalidates the session's publisher credential, requests publisher disconnection and cancels its jobs. Use this only for a temporary source. Requires video-inference-job:create. Use a workspace API key or OAuth access token for the workspace in the path.
- Write
Start a Workflow job
Starts metered processing. Send a new Idempotency-Key for one intentional run; retry uncertain responses with the same key, body and API key. Different keys can create separate billable jobs. A successful start can be queued or waiting, not yet running.
- Write
Stop a Workflow job
Requests an explicit stop. Persistent intent is disabled; queued jobs can be cancelled immediately. Active processors learn of cancellation through status updates, so the response is not a guarantee that shutdown has already finished.
- Write
Renew a temporary preview owner lease
Extends an active temporary-source preview's owner lease to at least two minutes from now, including while queued. It never revives an expired or stopped job and cannot extend the one-hour preview maximum.
- Write
Watch annotated Workflow output
Requests publication for at least 60 seconds on a claimed or running job. Renew at least every 30 seconds while viewing; renewal never shortens the lease. One image output is selected for the whole job, so another viewer's output selection changes every viewer's stream.
Workspaces and projects
- Read
Verify credentials and get the authenticated workspace
This operation is part of the public Roboflow Platform API.
- Read
Get a workspace and its projects
This operation is part of the public Roboflow Platform API.
- Read
Get a project and its versions
This operation is part of the public Roboflow Platform API.
- Read
Get dataset health metrics
This operation is part of the public Roboflow Platform API.
- Read
List models for a project
This operation is part of the public Roboflow Platform API.
- Read
Get a dataset version
This operation is part of the public Roboflow Platform API.
- Write
Create a project
This operation is part of the public Roboflow Platform API.
- Write
Fork a Universe project
This operation is part of the public Roboflow Platform API.
- Destructive
Move a project to Trash
This operation is part of the public Roboflow Platform API.
- Destructive
Move a dataset version to Trash
This operation is part of the public Roboflow Platform API.
API keys
- Read
List workspace API keys
This operation is part of the public Roboflow Platform API.
- Read
Get the workspace publishable API key
This operation is part of the public Roboflow Platform API.
- Read
Get an API key
This operation is part of the public Roboflow Platform API.
- Write
Create a workspace API key
This operation is part of the public Roboflow Platform API.
- Write
Update an API key
This operation is part of the public Roboflow Platform API.
- Destructive
Revoke an API key
This operation is part of the public Roboflow Platform API.
Project folders
- Read
List project folders
This operation is part of the public Roboflow Platform API.
- Write
Create a project folder
This operation is part of the public Roboflow Platform API.
- Write
Update a project folder
This operation is part of the public Roboflow Platform API.
- Write
Add or remove projects in a folder
This operation is part of the public Roboflow Platform API.
- Destructive
Delete a project folder
This operation is part of the public Roboflow Platform API.
- Destructive
Remove projects from a folder
This operation is part of the public Roboflow Platform API.
Dataset versions and training
- Read
Export a dataset version
This operation is part of the public Roboflow Platform API.
- Read
Get training results
This operation is part of the public Roboflow Platform API.
- Read
List training runs
This operation is part of the public Roboflow Platform API.
- Read
Get a training run
This operation is part of the public Roboflow Platform API.
- Read
Get the accepted training recipe
This operation is part of the public Roboflow Platform API.
- Write
Generate a dataset version
This operation is part of the public Roboflow Platform API.
- Write
Rebalance train, validation, and test splits
This operation is part of the public Roboflow Platform API.
- Write
Start a training job
This operation is part of the public Roboflow Platform API.
- Write
Stop an active training job
This operation is part of the public Roboflow Platform API.
- Write
Create a training run
This operation is part of the public Roboflow Platform API.
- Write
Stop an active training run
This operation is part of the public Roboflow Platform API.
- Destructive
Cancel a queued training job
This operation is part of the public Roboflow Platform API.
- Destructive
Cancel a queued training run
This operation is part of the public Roboflow Platform API.
- Destructive
Move a training run to Trash
This operation is part of the public Roboflow Platform API.
Images and batches
- Read
Get image details
This operation is part of the public Roboflow Platform API.
- Read
List image batches
This operation is part of the public Roboflow Platform API.
- Read
Get an image batch
This operation is part of the public Roboflow Platform API.
- Read
Get dataset ZIP upload status
This operation is part of the public Roboflow Platform API.
- Read
Get video upload status
Reads the processing status of a video upload.
- Write
Upload an image
This operation is part of the public Roboflow Platform API.
- Write
Upload an annotation
This operation is part of the public Roboflow Platform API.
- Write
Add or remove image tags
This operation is part of the public Roboflow Platform API.
- Write
Update metadata and tags for one image
This operation is part of the public Roboflow Platform API.
- Write
Update metadata and tags for multiple images
This operation is part of the public Roboflow Platform API.
- Write
Add Asset Library images to a project
Adds images or videos that are already in the Asset Library to a Project. Requires the image:create scope. API keys scoped to project folders cannot use it. The add runs as an async task.
- Write
Start a dataset ZIP upload
This operation is part of the public Roboflow Platform API.
- Write
Prepare a video upload
Creates a pending video and returns a signed URL for its file. The video joins the workspace, and the project when you set one, after you PUT the file to signedUrl and Roboflow processes it.
- Destructive
Save an image annotation with a version check
Replaces the project's annotation on one image. The save works only when expectedGroupVersion matches the current annotationGroupVersion, so it never overwrites an edit made by someone else. The API key needs the image:read and image:annotate scopes.
- Destructive
Delete images from a project
This operation is part of the public Roboflow Platform API.
Annotation
- Read
List annotation jobs
This operation is part of the public Roboflow Platform API.
- Read
Get an annotation job
This operation is part of the public Roboflow Platform API.
- Read
List Auto Label foundation models
Lists the Auto Label foundation models, with the credits each uses per image and whether the workspace plan includes it. Each model id is accepted as the modelType when you start a job or run a preview.
- Read
Preview an Auto Label model
Runs one image through an Auto Label foundation model and returns the predictions with a per-class summary. Previews are free and do not create a job.
- Read
Get Auto Label job status
This operation is part of the public Roboflow Platform API.
- Write
Add comments to an image
Adds comments to an image that is already in the project. Annotations do not change. Comments are attributed to the user of the API key. Requires the image:annotate scope and access to annotate the project. Video projects are not supported.
- Write
Create an annotation job
This operation is part of the public Roboflow Platform API.
- Write
Create an Auto Label job
This operation is part of the public Roboflow Platform API.
Workflows
- Read
List Workflows
This operation is part of the public Roboflow Platform API.
- Read
Get a Workflow
This operation is part of the public Roboflow Platform API.
- Read
List Workflow versions
This operation is part of the public Roboflow Platform API.
- Write
Create a Workflow
This operation is part of the public Roboflow Platform API.
- Write
Update a Workflow
This operation is part of the public Roboflow Platform API.
- Write
Fork a Workflow
This operation is part of the public Roboflow Platform API.
- Write
Create a Workflow token
This operation is part of the public Roboflow Platform API.
- Destructive
Move a Workflow to Trash
This operation is part of the public Roboflow Platform API.
Agent
- Read
List Roboflow Agent conversations
This operation is part of the public Roboflow Platform API.
- Read
Get a Roboflow Agent conversation
This operation is part of the public Roboflow Platform API.
- Write
Send a message to Roboflow Agent
This operation is part of the public Roboflow Platform API.
- Write
Publish a Workflow created by Roboflow Agent
This operation is part of the public Roboflow Platform API.
Search
- Read
Search public Universe projects
This operation is part of the public Roboflow Platform API.
- Read
Search public Universe projects by image
This operation is part of the public Roboflow Platform API.
- Read
Search images in a project
Searches the images in one project. An API key from the project's workspace can search any of its projects. An API key from another workspace can search public projects only.
- Read
Search images across a workspace
Searches the images across a workspace. The API key must belong to this workspace. Keys from other workspaces cannot search it, even when it has public projects.
- Read
Export image search results
This operation is part of the public Roboflow Platform API.
- Read
Get image search export status
This operation is part of the public Roboflow Platform API.
Trash and tasks
- Read
List deleted resources
This operation is part of the public Roboflow Platform API.
- Read
Get asynchronous task status
This operation is part of the public Roboflow Platform API.
- Write
Restore a deleted resource
This operation is part of the public Roboflow Platform API.
Deployment
- Read
Get project deployment settings
This operation is part of the public Roboflow Platform API.
- Read
Get Active Learning settings
This operation is part of the public Roboflow Platform API.
- Read
List Active Learning images
This operation is part of the public Roboflow Platform API.
- Write
Select the deployed project model
This operation is part of the public Roboflow Platform API.
- Write
Enable Active Learning
This operation is part of the public Roboflow Platform API.
- Write
Disable Active Learning
This operation is part of the public Roboflow Platform API.
- Write
Update Active Learning settings
This operation is part of the public Roboflow Platform API.
Monitoring
- Read
Get inference usage statistics
This operation is part of the public Roboflow Platform API.
- Write
Attach metadata to inference records
This operation is part of the public Roboflow Platform API.
Vision Events
- Read
Get Vision Event video status
This operation is part of the public Roboflow Platform API.
- Read
Get a Vision Event video playback URL
This operation is part of the public Roboflow Platform API.
- Read
Query Vision Events
This operation is part of the public Roboflow Platform API.
- Read
Get a custom metadata schema
This operation is part of the public Roboflow Platform API.
- Read
List Vision Event use cases
This operation is part of the public Roboflow Platform API.
- Write
Create a Vision Event
This operation is part of the public Roboflow Platform API.
- Write
Create Vision Events in a batch
This operation is part of the public Roboflow Platform API.
- Write
Upload a Vision Event image
This operation is part of the public Roboflow Platform API.
- Write
Prepare a Vision Event video upload
This operation is part of the public Roboflow Platform API.
- Write
Create a Vision Event use case
This operation is part of the public Roboflow Platform API.
- Write
Unarchive a Vision Event use case
This operation is part of the public Roboflow Platform API.
- Destructive
Delete Vision Events
This operation is part of the public Roboflow Platform API.
- Destructive
Update a Vision Event use case
This operation is part of the public Roboflow Platform API.
- Destructive
Archive a Vision Event use case
This operation is part of the public Roboflow Platform API.
Deployment Manager
- Read
List Deployment Manager devices
This operation is part of the public Roboflow Platform API.
- Read
Get the default device configuration
This operation is part of the public Roboflow Platform API.
- Read
Get a device
This operation is part of the public Roboflow Platform API.
- Read
Get device configuration
This operation is part of the public Roboflow Platform API.
- Read
Get device configuration history
This operation is part of the public Roboflow Platform API.
- Read
List device streams
This operation is part of the public Roboflow Platform API.
- Read
Get a device stream
This operation is part of the public Roboflow Platform API.
- Read
Get device logs
This operation is part of the public Roboflow Platform API.
- Read
Get device telemetry
This operation is part of the public Roboflow Platform API.
- Read
Get device events
This operation is part of the public Roboflow Platform API.
- Read
List device services
This operation is part of the public Roboflow Platform API.
- Read
Get a device service
This operation is part of the public Roboflow Platform API.
- Read
Get a device service metric series
This operation is part of the public Roboflow Platform API.
- Write
Register a Deployment Manager device
This operation is part of the public Roboflow Platform API.
Model evaluations
- Read
List model evaluations
This operation is part of the public Roboflow Platform API.
- Read
Compare model evaluations
Compare models on one Dataset Version. The operation includes models that have no evaluation. It is read-only and does not start evaluations.
- Read
Get a model evaluation
This operation is part of the public Roboflow Platform API.
- Read
Get mAP results
This operation is part of the public Roboflow Platform API.
- Read
Get a confidence sweep
This operation is part of the public Roboflow Platform API.
- Read
Get performance by class
This operation is part of the public Roboflow Platform API.
- Read
Get a confusion matrix
This operation is part of the public Roboflow Platform API.
- Read
Get vector analysis
This operation is part of the public Roboflow Platform API.
- Read
Get image predictions
This operation is part of the public Roboflow Platform API.
- Read
Get evaluation recommendations
This operation is part of the public Roboflow Platform API.
Billing
- Read
Get a billing usage report
This operation is part of the public Roboflow Platform API.
Metrics
- Read
Metrics
Endpoint that serves Prometheus metrics.
Info
- Read
Info
Get the server name and version number
Logs
- Read
Get Recent Logs
Get recent application logs for debugging
Model
- Read
Get model keys
Get the ID of each loaded model
Infer
- Write
Large multi-modal model infer
Run inference with the specified large multi-modal model
- Write
Large multi-modal model infer with model ID in path
Run inference with the specified large multi-modal model. Model ID is specified in the URL path (can contain slashes).
- Write
Depth Estimation
Run the depth estimation model to generate a depth map.
- Write
Depth Estimation with model ID in path
Run depth estimation. Model ID is specified in the URL path and can contain slashes.
Initialise webrtc worker
- Write
[EXPERIMENTAL] Establishes WebRTC peer connection and processes video stream in spawned process or modal function
[EXPERIMENTAL] Establishes WebRTC peer connection and processes video stream in spawned process or modal function
Webrtc
- Write
WebRTC session heartbeat
Receive heartbeat for an active WebRTC session. This endpoint is called periodically to indicate that their session is still active. The session will be removed from the quota count if no heartbeat is received within the TTL period. Requires apikey for authentication.
- Write
End WebRTC session
End a WebRTC session and immediately free the quota slot. Requires apikey for authentication.
Readiness
- Read
Readiness
Readiness endpoint for Kubernetes readiness probe.
Healthz
- Read
Healthz
Health endpoint for Kubernetes liveness probe. Verifies CUDA context health when running on GPU. Returns 503 if CUDA is corrupted (unrecoverable - requires process restart).
Clip
- Write
CLIP Image Embeddings
Run the Open AI CLIP model to embed image data.
- Write
CLIP Text Embeddings
Run the Open AI CLIP model to embed text data.
- Write
CLIP Compare
Run the Open AI CLIP model to compute similarity scores.
Perception encoder
- Write
PE Image Embeddings
Run the Meta Perception Encoder model to embed image data.
- Write
Perception Encoder Text Embeddings
Run the Meta Perception Encoder model to embed text data.
- Write
Perception Encoder Compare
Run the Meta Perception Encoder model to compute similarity scores.
Grounding dino
- Write
Grounding DINO inference.
Run the Grounding DINO zero-shot object detection model.
Yolo world
- Write
YOLO-World inference.
Run the YOLO-World zero-shot object detection model.
Doctr
- Write
DocTR OCR response
Run the DocTR OCR model to retrieve text in an image.
Easy ocr
- Write
EasyOCR OCR response
Run the EasyOCR model to retrieve text in an image.
Sam
- Write
SAM Image Embeddings
Run the Meta AI Segmant Anything Model to embed image data.
- Write
SAM Image Segmentation
Run the Meta AI Segmant Anything Model to generate segmenations for image data.
Sam2
- Write
SAM2 Image Embeddings
Run the Meta AI Segment Anything 2 Model to embed image data.
- Write
SAM2 Image Segmentation
Run the Meta AI Segment Anything 2 Model to generate segmenations for image data.
Sam3
- Write
Seg preview Image Embeddings
Run the Model to embed image data.
- Write
SAM3 PCS (promptable concept segmentation)
Run the SAM3 PCS (promptable concept segmentation) to generate segmentations for image data.
- Write
SAM3 PVS (promptable visual segmentation)
Run the SAM3 PVS (promptable visual segmentation) to generate segmentations for image data.
Sam3 3d
- Write
SAM3 3D Object Generation
Generate 3D meshes and Gaussian splatting from 2D images with mask prompts.
Owlv2
- Write
Owlv2 image prompting
Run the google owlv2 model to few-shot object detect
Ocr
- Write
TrOCR OCR response
Run the TrOCR model to retrieve text in an image.
- Write
PP-OCRv6 OCR response
Run PP-OCRv6 two-stage OCR to retrieve text in an image.
General
- Read
Legacy Infer From Request
Legacy inference endpoint for object detection, instance segmentation, and classification.
- Write
Legacy Infer From Request
Legacy inference endpoint for object detection, instance segmentation, and classification.
Dashboard.html
- Read
Dashboard Guard
GET /dashboard.html
- Read
Dashboard Guard
HEAD /dashboard.html
No operations match.
See How connections work for access and permissions, or browse all integrations.