Imagga Image Recognition API

Name: Imagga Image Recognition API
Brand: Imagga Image Recognition API
Availability: InStock

★ Only Publicly Available OpenAPI DocumentAI/MLVisionbasic11 EndpointsREST

For Agents

Tag, categorise, extract colors from, detect faces in, crop, and run OCR on images using Imagga's REST endpoints with a shared upload ID.

Install Jentic One Beta

Connect the Imagga Image Recognition API to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Imagga Image Recognition API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.

Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.

Step 1: Jentic One Host machine

# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh

Step 2: Agent machine

# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register       # connects your agent to your Jentic One instance

Jentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.

Capabilities

What an agent can do with Imagga Image Recognition API.

Auto-tag an uploaded image with confidence scores per tag

Classify an image against a chosen categorizer such as nsfw or personal photos

Extract dominant colors and palettes from an image

GET STARTED

Start building with Imagga Image Recognition API

Explore with Jentic

View OpenAPI Document

Use Cases

Patterns agents use Imagga Image Recognition API for, with concrete tasks.

★ E-Commerce Catalogue Auto-Tagging

Online stores often have thousands of product photos with sparse metadata. Pushing each image to /uploads, calling /tags with the upload ID, and storing the high-confidence tags back on the product record produces a searchable catalogue without manual tagging. The shared upload ID means a single image can also feed /colors and /croppings without re-uploading.

POST /uploads with the product image, GET /tags with the upload_id, then write the high-confidence tags back to the product record.

Content Moderation for User Uploads

Social and dating apps need to screen user-uploaded images for NSFW content and unsafe imagery. The /categories/{categorizer_id} endpoint with the nsfw categorizer returns class confidence scores that a moderation queue can act on, blocking or flagging content above a threshold.

Upload the user image, call /categories/nsfw_beta with the upload_id, and queue the image for human review if the unsafe class confidence is above 0.7.

Smart Cropping for Responsive Images

Editorial sites need to render hero images at multiple aspect ratios without cutting off the subject. /croppings returns coordinates that preserve the focal point, so a CDN or image proxy can produce 16:9, 1:1, and 4:5 variants from one source image. Combined with /colors, the same upload also yields a placeholder color for skeleton screens.

Upload a hero image, call /croppings for each target aspect ratio, and emit the crop coordinates to the image proxy that serves the front page.

OCR for Receipt Capture

Expense apps need to pull text out of photographed receipts. /text returns recognised text from the supplied image, which a downstream parser can map to fields like merchant, total, and date. Pairing this with face detection lets the same pipeline blur faces in any background people accidentally captured.

Upload the receipt photo, call /text with the upload_id, and pass the recognised text to the receipt parser.

Agent-Driven Vision Tasks

An AI agent that acts on a user's photo (tag, moderate, crop, extract text) can search Jentic for the right Imagga operation and run it with the upload ID it already holds, rather than chaining HTTP calls by hand. Through Jentic, basic auth credentials stay vaulted and the agent only sees the upload reference.

Search Jentic for 'tag an image with Imagga', load the /tags schema, and execute it with the upload ID returned from a prior /uploads call.

Key Endpoints

11 endpoints — jentic publishes the only available openapi specification for imagga image recognition api, keeping it validated and agent-ready.

METHOD

PATH

DESCRIPTION

POST

/uploads

Upload an image and receive an upload ID

GET

/tags

Auto-tag an uploaded image

GET

/categories/{categorizer_id}

Classify against a named categorizer

GET

/colors

Extract dominant colors

GET

/faces/detections

Detect faces in an image

GET

/croppings

Compute smart crop coordinates

GET

/text

Run OCR on an image

GET

/usage

Check API usage and remaining quota

POST

/uploads

Upload an image and receive an upload ID

GET

/tags

Auto-tag an uploaded image

GET

/categories/{categorizer_id}

Classify against a named categorizer

GET

/colors

Extract dominant colors

GET

/faces/detections

Detect faces in an image

Why Jentic?

Three things that make agents converge on Jentic-routed access.

Credential isolation

Your Imagga API key and secret are stored once, encrypted, by your own Jentic One instance, and the Basic auth header is generated at execution time. They never enter the agent's prompt, logs, or context.

Intent-based discovery

Agents search Jentic by intent such as 'tag an image' or 'detect faces', and Jentic returns the matching Imagga operation with its query schema, including the upload_id that ties analysis to an uploaded asset, so the agent calls the right endpoint without browsing the reference docs.

Time to first call

FAQs

Specific to using Imagga Image Recognition API through Jentic.

Why is there no official OpenAPI spec for Imagga Image Recognition API?

Imagga publishes documentation but no canonical OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Imagga via 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 Imagga API use?

The API uses HTTP Basic auth with an API key and secret as the username and password pair. Through Jentic the secret is held in the encrypted vault and the Authorization header is built at execute time.

Can I run multiple analyses on a single uploaded image?

Yes. POST /uploads once and reuse the returned upload_id with /tags, /categories/{categorizer_id}, /colors, /faces/detections, /croppings, and /text without uploading again.

Does the API support content moderation?

Yes. GET /categories/{categorizer_id} with a moderation categorizer returns class confidence scores that a moderation queue can use to flag or block uploads above a chosen threshold.

How do I tag an image through Jentic?

Run pip install jentic, search for 'tag an image with Imagga', load the /tags schema, and execute with the upload_id from a prior /uploads call. The agent receives the high-confidence tags as a structured list.

How do I check my remaining Imagga quota?

GET /usage returns request counts and remaining monthly quota for the authenticated account, which is useful for back-pressure logic in batch jobs.

Can I limit what my agent is allowed to do with the Imagga Image Recognition API?

Yes. Because Jentic One is self-hosted, you decide which Imagga operations the agent may call and which credentials it may use, and Imagga takes the image or upload reference in the request rather than as a path resource id, so you scope by operation. You might allow an agent to upload images, auto-tag them via GET /tags, and extract colors via GET /colors, while withholding face detection at GET /faces/detections, NSFW classification at GET /categories/{categorizer_id}, and OCR at GET /text. An agent limited to tagging never gets face detection or OCR unless you add those operations to its allowed set.