5 APIs across 3 product groups.
Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Hirak APIs, 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.
What agents get from Jentic-routed access to this vendor.
Setup
Wiring these APIs by hand means handling three different auth styles (RapidAPI headers, a Hirak X-API-Key, and no credential) and routing each task to the right endpoint. Through Jentic you install Jentic One once, add the Hirak APIs you need from the Jentic API Directory, store each key once, and your agent calls them through one execution path.
Permission scoping
These APIs are read-only or task-scoped and select inputs through query parameters, path values, or request bodies rather than scopable account resources, so scoping stays at the operation level. Your own rules in your Jentic One instance decide which operations, such as currency conversion, OCR, or face comparison, the agent may run.
Credential isolation
Specific to using Hirak APIs through Jentic.
What kinds of APIs does Hirak offer?
Hirak provides several independent utility APIs: foreign-exchange and crypto rates with conversion, face detection and analysis, text translation and language detection, image OCR in more than 100 languages, and IP-to-country geolocation. Each is a separate service with its own small endpoint surface.
Do these APIs share a single account or key?
No. The Exchange Rates and FaceAPI services are distributed through RapidAPI and use RapidAPI headers, the OCR and Translation APIs use a Hirak X-API-Key, and the IP to Country API needs no credential. You obtain and manage credentials per API rather than through one shared account.
Why does Jentic publish OpenAPI specs for Hirak?
Most Hirak APIs do not ship an official OpenAPI specification. Jentic generates and maintains these specs, validated against the live APIs, so agents and developers can call them through structured tooling. The IP to Country spec is vendor-official.
Are these APIs suitable for high-volume production workloads?
They are designed as lightweight, task-specific services rather than platforms with published SLAs. They suit prototypes, low-to-moderate volume features, and agent workflows. For strict throughput or uptime guarantees, pair them with or move to a provider that publishes those terms.
BOOK A DEMO
Browse thousands of APIs and connect them all to your agent with Jentic One. One layer, one credential — every API your agent needs.
For Agents
An agent can convert currencies and read FX or crypto rates, detect and compare faces or estimate age and emotion, translate text and detect its language, extract text from images in more than 100 languages, and resolve an IP address to a country. Each capability is a distinct Hirak API with a small endpoint set the agent selects by intent.
Hirak groups several unrelated utility APIs under one provider, each kept intentionally narrow rather than bundled into one large service. This suits agents that need one specific task done at low cost without the integration weight of a broad cloud-vision, translation, or FX platform.
Use for: Lightweight, task-specific lookups: currency and crypto conversion, face detection and analysis, text translation and language detection, image OCR, and IP-to-country resolution through small header-authenticated or public APIs.
Not supported: payment execution, full KYC verification, city-level geolocation, structured document extraction, speech translation, real-time streaming
Credentials: Authentication is per API rather than shared: the FaceAPI and Exchange Rates APIs use RapidAPI headers, the OCR and Translation APIs use a Hirak X-API-Key header, and the IP to Country API needs no credential, so one key does not cover the whole portfolio.
Across 5 Hirak APIs: 1 vendor-official, 4 Jentic-generated, all indexed by Jentic and kept validated and agent-ready.
Hirak is an independent provider of small, single-purpose web APIs covering foreign-exchange and crypto rates, face detection and analysis, text translation and language detection, image OCR, and IP-to-country geolocation. Each API keeps a narrow endpoint surface, so integration is fast and the request shape is easy to reason about. Most APIs authenticate with a header key (RapidAPI or a Hirak X-API-Key), while the IP-to-country lookup is unauthenticated. Jentic publishes and maintains the OpenAPI specifications for these APIs so agents and developers can call them through structured tooling.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fhirak.site" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fhirak.site" | sh
jentic register # connects your agent to your Jentic One instanceJentic 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.
Once connected, ask your agent something like: “convert currency, translate text, or extract text from an image with a Hirak API”.
Each workflow spans multiple Hirak APIs. Jentic routes each operation to the right API automatically.
Localised multilingual storefront
An agent reads a visitor's country from their IP to default the currency, then converts base prices into the local currency and translates product copy or reviews into the local language, so a store renders localised without maintaining separate rate and translation jobs.
Call GET / on the IP to Country API for the visitor country, GET /convert on the Exchange Rates API to localise the base price, and POST /translate on the Translation API to render product copy in the visitor language.
Hirak IP to Country API + Hirak Exchange Rates API + Hirak Translation API
Multilingual document reader
An agent extracts text from a scanned or photographed document with OCR, detects the language of the extracted text, then translates it into the reader's language, turning an image of a foreign-language document into readable text.
Call POST /ocr on the OCR API to get the text, POST /detect on the Translation API to identify the source language, then POST /translate to convert it into the target language.
Hirak OCR API + Hirak Translation API
Identity check with geo signal
An onboarding agent compares a selfie against an ID photo for a similarity score and analyses the face for an age estimate, while resolving the sign-up IP to a country to flag mismatches with the stated region before routing to manual review.
Call POST /compare and POST /analyze on the FaceAPI for a similarity score and age estimate, and GET /{ip} on the IP to Country API to compare the sign-up country against the declared region.
Hirak FaceAPI + Hirak IP to Country API
Each Hirak credential you use, whether a RapidAPI key or an X-API-Key, is stored once and encrypted by your own Jentic One instance and injected as the right header at execution time. The unauthenticated IP lookup has no key to store. No secret enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search the Jentic directory by intent, such as convert currency, translate text, or extract text from an image, and Jentic returns the matching Hirak operation with its schema so the agent picks the correct endpoint across the portfolio without reading reference docs.
Can an agent chain several Hirak APIs together?
Yes. Because each API is narrow and predictable, an agent can combine them, for example OCR then translate a document, or resolve a country then convert prices and translate copy, treating each as a separate tool in one workflow.
How does an agent pick the right Hirak API for a task?
An agent searches Jentic by intent, such as convert currency, extract text from an image, or compare two faces, and Jentic returns the matching operation with its schema so the agent selects the correct endpoint without reading reference docs.
All 4 are in the Jentic catalogue with the same one-credential, intent-search pattern.