Product

How Jentic Works

From API assessment to production deployment in four simple steps.

Product Overview

PLATFORM

Jentic OneSelf-hosted, open-source control plane between your agents and any APIAPI DirectoryBrowse 10,000+ APIs ready for AI agent integrationAPI ScorecardAssess your APIs for AI-readiness with automated scoringAgentic SandboxSafely simulate AI agents with your production APIsJenticSign in to the Jentic web app

CAPABILITIES

IntegrationConnect AI agents to your existing systemsWorkflowsDiscover and capture successful agent workflowsGovernanceDefine, observe, and enforce AI policies

TOOLS

Arazzo UIVisualize Arazzo workflows as interactive documentationArazzo EditorBuild and edit multi-step API workflows visually
Pricing
Developers

GET STARTED

DocumentationGuides and API referenceQuickstartGet up and running in minutes

COMMUNITY

GitHubOpen source projects and examples
Resources
BlogLatest articles and insightsPress & MediaBrand assets and press contactOpen StandardsBuilt on open specs. Never locked in.NewsletterAPIs, AI agents, mixed with architecture and strategy.
Company
About UsOur mission and teamCareersJoin our teamContactGet in touch
Install Jentic OneBook a Demo
How Jentic WorksJentic OneAPI DirectoryAPI ScorecardAgentic SandboxJenticIntegrationWorkflowsGovernanceArazzo UIArazzo Editor
Pricing
DocumentationQuickstartGitHub
BlogPress & MediaOpen StandardsNewsletter
About UsCareersContact
Request a demoInstall Jentic One
Jentic
For Enterprises
  • Product Overview
  • Agentic Sandbox
  • Book a Demo
For Developers
  • Jentic One
  • Documentation
  • GitHub
Company
  • About Jentic
  • Careers
  • Contact Us
ISO/IEC 27001:2022 certification badge issued by Prescient Security

Information Security Management System

Certified to ISO/IEC 27001:2022 by Prescient Security

Terms & Conditions•Privacy Policy•
© 2026 Jentic. All rights reserved.
APIs / AI/ML / EyePop WebAPI
EyePop WebAPI logo

EyePop WebAPI

★ Only Publicly Available OpenAPI DocumentAI/MLVisionoauth2, bearer104 EndpointsREST

For Agents

Create and run EyePop computer-vision Pops, manage models and inputs, and administer accounts, API keys and resource pools through 104 endpoints under api.eyepop.ai.

Use for: I need to create a new EyePop Pop for object detection, Start an existing Pop and run inference on an uploaded image, Attach a vision model to an existing Pop, Reserve compute resources for a vision workload

Not supported: Does not handle text generation, audio transcription, or training of new vision models — use for orchestrating EyePop computer-vision pipelines and account administration only.

Jentic publishes the only available OpenAPI specification for EyePop WebAPI, keeping it validated and agent-ready. The EyePop WebAPI is the control plane for the EyePop computer-vision platform, letting customers manage Pops (vision pipelines), upload inputs, attach models, run inference, and administer accounts, API keys and resource pools. Authentication is by Auth0 OAuth 2.0 implicit flow for interactive users or by JWT bearer tokens issued through the API-key endpoints for server-side workloads.

Jentic One on GithubView OpenAPI Document

Install Jentic One Beta

Connect the EyePop WebAPI to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the EyePop WebAPI, 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.

1

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
2

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 EyePop WebAPI API.

Create, configure and run vision Pops via /api/v1/user/pops to execute computer-vision pipelines

Attach models to a Pop with POST /api/v1/user/pops/{id}/models to compose multi-stage inference

Reserve and manage compute resources through /api/v1/resources/reserve and the admin resource endpoints

Issue, list and revoke API keys for server-side access via /api/v1/api-keys

Manage user accounts and team membership through /api/v1/accounts and /api/v1/accounts/{account_uuid}/users

Inspect user credits, plans and usage with /api/v1/user/credits and /api/v1/user/me

Use Cases

Patterns agents use EyePop WebAPI API for, with concrete tasks.

★ Build and Run a Vision Pipeline

Create an EyePop Pop, attach the required models, and run inference against uploaded inputs through a single REST surface. Pops are the orchestration unit for EyePop's vision pipelines and expose endpoints for start, stop, recreate, copy, and configuration retrieval. Iteration is fast because Pop edits and model attachments propagate without redeploying client code.

POST /api/v1/user/pops to create a Pop, POST /api/v1/user/pops/{id}/models to attach a detection model, then GET /api/v1/user/pops/{id}/start to run inference.

Server-Side Integration via API Keys

Provision long-lived API keys for backend services that need to call EyePop without going through Auth0. The /api/v1/api-keys endpoints create, list, update and delete keys, and the resulting JWT bearer tokens are accepted by all user-facing operations on the API. This is the recommended path for cron jobs and webhook handlers that ingest media at scale.

POST /api/v1/api-keys to issue a key, store the resulting JWT in the secret manager, and use it as the Bearer token on subsequent calls to /api/v1/user/pops/{id}/start.

Account and Resource Administration

Administer EyePop accounts, members, and reserved compute pools through the admin and account endpoints. Account owners can list, create, update and delete accounts, manage team membership at /api/v1/accounts/{account_uuid}/users/{user_uuid}, reserve resources at /api/v1/resources/admin/resource/reserve, and inspect resource pools and usage. Useful for platform engineers operating EyePop at organisational scale.

POST /api/v1/accounts to create the account, POST /api/v1/accounts/{account_uuid}/users/{user_uuid} to add a teammate, then POST /api/v1/resources/admin/resource/reserve to lock a GPU for the account's workload.

Agent-Driven Vision Workflows via Jentic

An AI agent that needs computer vision can use Jentic to discover EyePop Pop operations, attach models, and run inference without hand-coding the client. The agent searches by intent, loads the operation schema, and the API key or Auth0 token stays inside the Jentic vault rather than being exposed in the model context.

Through Jentic, search 'run an EyePop computer-vision pipeline', load POST /api/v1/user/pops and the start operation, then execute them with the configured model id.

Key Endpoints

104 endpoints — jentic publishes the only available openapi specification for eyepop webapi, keeping it validated and agent-ready.

METHOD

PATH

DESCRIPTION

POST

/api/v1/user/pops

Create a new Pop

GET

/api/v1/user/pops/{id}/start

Start a Pop

GET

/api/v1/user/pops/{id}/stop

Stop a Pop

POST

/api/v1/user/pops/{id}/models

Attach a model to a Pop

POST

/api/v1/api-keys

Issue a new API key

GET

/api/v1/user/credits

Get the current user's credit balance

POST

/api/v1/accounts

Create an EyePop account

POST

/api/v1/resources/admin/resource/reserve

Reserve compute resource

POST

/api/v1/user/pops

Create a new Pop

GET

/api/v1/user/pops/{id}/start

Start a Pop

GET

/api/v1/user/pops/{id}/stop

Stop a Pop

POST

/api/v1/user/pops/{id}/models

Attach a model to a Pop

POST

/api/v1/api-keys

Issue a new API key

GET

/api/v1/user/credits

Get the current user's credit balance

POST

/api/v1/accounts

Create an EyePop account

POST

/api/v1/resources/admin/resource/reserve

Reserve compute resource

Why Jentic?

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

Credential management

Credential isolation

Auth0 access tokens and EyePop JWT bearer tokens issued by /api/v1/api-keys are stored encrypted in the Jentic vault; the agent receives a scoped token at execution time and never handles the long-lived secret.

Intent-based discovery

Intent-based discovery

Agents search by intent — for example 'create an EyePop Pop' or 'reserve a GPU resource' — and Jentic returns the matching operation among the 104 endpoints with its full input schema.

Time to first call

Time to first call

Direct integration: 2-5 days to wire Auth0, model attachment, and Pop lifecycle handling. Through Jentic: under an hour to discover, load, and execute the first Pop call.

Related APIs

Alternatives and complements available in the Jentic catalogue.

Complementary

Exude API Service

→

Lightweight English text preprocessing — useful when EyePop pipelines emit captions or transcripts that need cleanup.

Use Exude after EyePop has produced text outputs that need stop-word removal or stemming before downstream search.

Complementary

Eyefinity Optometry EHR FHIR API

→

Healthcare FHIR reads — pair with EyePop when building optometry-imaging workflows that combine vision inference and clinical context.

Combine with Eyefinity FHIR when the agent needs both computer-vision results and the patient's chart to drive a clinical decision.

FAQs

Specific to using EyePop WebAPI API through Jentic.

Why is there no official OpenAPI spec for EyePop WebAPI?

EyePop does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call EyePop WebAPI via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.

What authentication does the EyePop WebAPI use?

The spec declares two security schemes: an Auth0 OAuth 2.0 implicit flow at https://auth0.eyepop.ai/authorize for interactive users, and HTTP Bearer JWT tokens for server-side calls. Tokens are stored in the Jentic vault and injected at execution time so they never appear in the agent prompt.

Can I run object detection through the EyePop WebAPI?

Yes. Create a Pop with POST /api/v1/user/pops, attach a vision model with POST /api/v1/user/pops/{id}/models, and start the pipeline with GET /api/v1/user/pops/{id}/start to run inference on the configured inputs.

How do I issue a server-side API key?

Call POST /api/v1/api-keys to create a key, then list and revoke keys via GET /api/v1/api-keys and DELETE /api/v1/api-keys/{id}; the issued JWT is the Bearer token used for headless calls.

How do I run a Pop through Jentic?

Run pip install jentic, search 'run an EyePop computer-vision pipeline', load the POST /api/v1/user/pops and start operations, then execute them with the configured model id; Jentic injects the bearer token from the vault.

What are the rate limits for the EyePop WebAPI?

The spec does not enumerate hard rate limits, so honour 429 responses with exponential backoff and check your account credit balance via GET /api/v1/user/credits before launching large batch runs.

GET STARTED

Start building with EyePop WebAPI API

Explore with Jentic
View OpenAPI Document