canonical: https://jentic.com/apis/claid.ai/claid

# Claid.ai Image Editing API

Jentic publishes the only available OpenAPI specification for Claid.ai Image Editing API, keeping it validated and agent-ready. Claid.ai applies AI-powered image edits - enhancement, upscaling, background removal, smart cropping, and format conversion - at scale through a single REST surface. The API offers synchronous, asynchronous, and batch processing modes, plus direct URL-based upload and persistent storage connectors so output can be written back to S3, GCS, or other targets without a round-trip through the caller. Authentication is by bearer token using a Claid.ai API key.

## For AI agents

Run AI image edits - upscale, remove background, smart crop, convert formats - synchronously, asynchronously, or in batch through Claid.ai. Useful for any agent that needs to produce ecommerce-ready or social-ready images on demand.

## Scope

Does not handle image classification, OCR, or video editing - use for AI image editing operations like upscaling, background removal, smart cropping, and format conversion only.

## Capabilities

- Apply AI image edits such as upscaling, background removal, and smart cropping in a single synchronous call
- Submit large or slow edit jobs asynchronously and poll the job id for completion
- Process many images in one batch and track the batch id through to completion
- Upload an image directly to Claid for editing without first hosting it elsewhere
- Configure persistent storage connectors so edited output is written to your own S3 or cloud bucket
- Convert image format and adjust output quality alongside the visual edits in the same request

## Use cases

### Ecommerce product image pipeline

An online retailer feeds raw product photos through Claid to remove backgrounds, smart-crop to a consistent aspect ratio, and upscale for the product listing page. Large drops are submitted via POST /v1-beta1/image/edit/batch and the resulting batch id is polled with GET /v1-beta1/image/edit/batch/{batchId}. With a storage connector configured, the cleaned images land directly in the retailer's CDN bucket.

Example prompt: Submit a batch of 50 product image URLs to /v1-beta1/image/edit/batch with background removal and 1:1 smart crop, then poll the batch id every 30 seconds

### On-demand social media image preparation

A social-publishing tool calls POST /v1-beta1/image/edit synchronously to crop and resize a hero image to multiple platform-specific dimensions in one request. The synchronous mode keeps the user-facing flow under a few seconds for typical input sizes, while async is reserved for heavier upscales. The bearer-token auth lets the publishing tool call Claid as a backend service without exposing the key to the browser.

Example prompt: POST /v1-beta1/image/edit with a source URL and a smart-crop preset for 1080x1350, then return the result URL

### Storage connector setup for private buckets

Teams that do not want to relay images through their own infrastructure configure a storage connector via POST /v1-beta1/storage/connectors. Claid then writes edit outputs straight to the configured bucket and the team manages credentials in one place rather than per-request. GET, PUT, and DELETE on /v1-beta1/storage/connectors/{connectorId} cover the lifecycle.

Example prompt: Create an S3 storage connector for bucket 'product-images-prod' with the supplied access key, then verify it appears in GET /v1-beta1/storage/connectors

### Agent integration via Jentic

Through Jentic, an agent helping a marketing team can resolve intents like 'upscale this image to 4K' to the right Claid endpoint and parameters without browsing the docs. The bearer API key is held in the Jentic credential vault and injected only at execution time.

Example prompt: Search Jentic for 'remove image background', load the schema for POST /v1-beta1/image/edit, and execute it with the user-supplied source URL

## Key endpoints

| Method | Path | Description |
| --- | --- | --- |
| POST | /v1-beta1/image/edit | Edit an image synchronously |
| POST | /v1-beta1/image/edit/async | Submit an async edit job |
| GET | /v1-beta1/image/edit/async/{jobId} | Check async job status |
| POST | /v1-beta1/image/edit/batch | Submit a batch edit |
| GET | /v1-beta1/image/edit/batch/{batchId} | Check batch status |
| POST | /v1-beta1/image/edit/upload | Upload an image and edit it |
| POST | /v1-beta1/storage/connectors | Create a storage connector |
| GET | /v1-beta1/storage/connectors | List storage connectors |

## Key resources

- **Image Editing** — Synchronous edit endpoint for single-image transformations
- **Async** — Async edit submission and job-status polling
- **Batch** — Multi-image batch edits with batch-id tracking
- **Upload** — Direct upload-and-edit flow without external hosting
- **Storage** — Configure persistent storage connectors that receive edit outputs

## Why Jentic

- **Setup:** Wiring the Claid.ai Image Editing API by hand means learning its bearer token auth and handling synchronous edits, async jobs, and batch polling yourself. Through Jentic you install once, import the Claid.ai Image Editing API from the API Directory, store the token once, and your agent calls it.
- **Permission scoping:** Claid takes the image and edit parameters in the request body rather than the URL path, so scope the agent to the operations it needs, such as running an image edit and polling its async job. You choose the operations it may call, so batch editing or storage-connector setup is not included unless you add it.
- **Credential handling:** Your Claid token is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
- **Discovery method:** Agents search Jentic by intent such as 'upscale an image' or 'remove an image background', and Jentic returns the matching Claid operation with its input schema so the agent calls the right endpoint without browsing the reference docs.

## Related APIs

- **Clarifai API** — Clarifai classifies and labels images that Claid then edits or prepares for delivery.
- **Clarifai** — Clarifai's image inputs API overlaps for tagging and content-moderation, though it does not perform pixel-level edits.
- **Civic Auth API** — Civic Auth gates the user identity that often sits in front of an image-upload pipeline.

## FAQ

### Why is there no official OpenAPI spec for Claid.ai Image Editing API?

Claid.ai does not publish an OpenAPI specification. Jentic generates and maintains this spec from the Claid developer documentation so AI agents and developers can call Claid through structured tooling. It is validated against the live API and kept up to date. Get started with Jentic One, the self-hosted execution layer.

### What authentication does the Claid.ai API use?

Claid.ai uses HTTP bearer authentication - your Claid API key is sent as the bearer token in the Authorization header. Through Jentic, the key is stored encrypted in the credential vault and injected only at execution time.

### Can I process many images in one call with Claid.ai?

Yes. POST /v1-beta1/image/edit/batch accepts a list of source images and a transformation set, and returns a batchId. Poll GET /v1-beta1/image/edit/batch/{batchId} for completion. For single heavy edits, the async endpoint at /v1-beta1/image/edit/async returns a jobId you poll the same way.

### How do I remove a background through Jentic?

Search Jentic for 'remove image background'. Jentic returns POST /v1-beta1/image/edit with the input schema for transformations. Load the schema, set the background-removal flag in the operations payload, and execute it with the source image URL - no need to read the Claid docs first.

### What are the rate limits for the Claid.ai API?

The OpenAPI spec does not declare formal rate limits. Claid applies per-account quotas tied to the plan; treat 429 responses as a signal to back off and confirm sustained throughput needs with Claid support before peak load.

### Can Claid.ai write outputs directly to my own cloud storage?

Yes. Configure a storage connector via POST /v1-beta1/storage/connectors with credentials for an S3 or compatible bucket. Subsequent edit jobs can target that connector so Claid writes the result directly to your bucket without a round-trip through your servers.

### Can I limit what my agent is allowed to do with the Claid.ai Image Editing API?

Yes. Because you run Jentic One yourself, you decide which Claid.ai operations the agent may call and hold the API token, so you can allow just an image edit at POST /v1-beta1/image/edit and polling its async job while leaving everything else out. Claid.ai takes the image and edit parameters in the request body rather than the URL path, so scoping is done at the operation level rather than by path patterns. If you do not grant them, batch editing at /v1-beta1/image/edit/batch and storage-connector setup at /v1-beta1/storage/connectors stay off limits to the agent.