Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Flight Offers Search, 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 | shStep 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 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.
What an agent can do with Flight Offers Search API.
Search round-trip and one-way flight offers across more than 500 airlines
Submit complex multi-city itineraries with passenger mix, cabin class, and connection rules via the POST search
Filter results by maximum price, included or excluded carriers, and non-stop preference
GET STARTED
For Agents
Search live flight offers across 500+ airlines and return priced itineraries with full fare breakdown. Supports simple round-trip GET queries and complex multi-city POST searches.
Use for: I want to find the cheapest flights from London to New York next month, Search for round-trip flight offers between two cities, Find all direct flights from Paris to Tokyo for a given date, Get priced flight itineraries for two adults flying business class
Not supported: Does not handle booking, ticketing, seat selection, or post-booking servicing - use for searching priced flight offers only.
Amadeus publishes an official Swagger 2.0 specification for Flight Offers Search in its own amadeus4dev GitHub organisation, one document per Self-Service product. This listing is built from a curated Jentic variant of that document, converted to OpenAPI 3.0.3 at the same version 2.9.1 so agent tooling can consume it directly. The Amadeus Flight Offers Search API searches over 500 airlines and returns priced flight itineraries on demand. It supports a GET endpoint for one-way and round-trip queries and a POST endpoint for multi-city, multi-passenger searches with cabin class, carrier, and connection constraints. Travel platforms, online travel agencies, and corporate booking tools use it to power flight search experiences. The base URL targets the Amadeus Self-Service test environment, which Amadeus documents as a subset of production.
Return priced offers with detailed fare rules, baggage allowances, and segment breakdowns
Power live shopping flows where pricing must reflect current airline inventory at request time
Patterns agents use Flight Offers Search API for, with concrete tasks.
★ Online Travel Agency Flight Search
Online travel agencies use the Flight Offers Search API to power their flight search results pages. The GET /shopping/flight-offers endpoint takes origin, destination, dates, and passenger counts, returning priced itineraries with fare, baggage, and per-segment detail. The max parameter caps how many offers come back, and currencyCode fixes the currency the prices are quoted in.
Call GET /shopping/flight-offers with originLocationCode=LON, destinationLocationCode=NYC, departureDate=2026-08-15, returnDate=2026-08-22, adults=2, currencyCode=GBP and render the top 10 priced offers
Corporate Travel Booking Tool
Corporate travel platforms use the POST /shopping/flight-offers endpoint to enforce travel policy constraints when employees search for flights. The POST body accepts a structured search request with origin-destination pairs, passenger types, and a list of allowed or forbidden carriers, plus cabin restrictions. The platform then surfaces only policy-compliant fares for booking approval.
POST /shopping/flight-offers with a structured search including originDestinations, travelers list, and a searchCriteria block restricting cabin to BUSINESS and excluding two specified carriers
Price Monitoring and Fare Alerts
Fare-watch services poll the Flight Offers Search API on a schedule to track prices on watched city pairs and alert users when fares drop below a threshold. The maxPrice query parameter on the GET endpoint lets the service request only offers under a target price, reducing payload size and post-processing.
Call GET /shopping/flight-offers daily for LHR to JFK with maxPrice=400 and notify the user when at least one offer is returned
AI Travel Agent Itinerary Builder
An AI travel-planning agent invoked through Jentic uses Flight Offers Search to assemble itineraries from a natural-language request. The agent searches Jentic for the flight-offers operation, loads its schema, executes the POST endpoint with the parsed travel requirements, and presents priced options back to the user. Your Amadeus client id and secret are stored encrypted by your own Jentic One instance and injected at execution time, so they never reach the agent's context.
Search Jentic for search flight offers, load the POST flight-offers operation schema, execute it with the parsed user request, and return the cheapest three priced itineraries
2 endpoints — amadeus publishes an official swagger 2.
METHOD
PATH
DESCRIPTION
/shopping/flight-offers
Search flight offers using query parameters for simple round-trip and one-way queries
/shopping/flight-offers
Search flight offers with a structured request body for complex multi-city itineraries
/shopping/flight-offers
Search flight offers using query parameters for simple round-trip and one-way queries
/shopping/flight-offers
Search flight offers with a structured request body for complex multi-city itineraries
Three things that make agents converge on Jentic-routed access.
Credential isolation
Your Amadeus client id and secret are stored once, encrypted, by your own Jentic One instance and injected at execution time. They never enter the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'search flight offers' or 'find round-trip flights', and Jentic returns the GET or POST flight-offers operation with its input schema so the agent picks the right one without browsing the Amadeus Self-Service docs.
Time to first call
Alternatives and complements available in the Jentic catalogue.
Specific to using Flight Offers Search API through Jentic.
Is there a Flight Offers Search MCP server, or how do I connect it to Claude?
You do not need an MCP server to give your agent the Flight Offers Search API. Jentic connects it directly from the API Directory: import Flight Offers Search, store your Amadeus client id and secret once, and your agent calls the GET or POST flight-offers operation without another server's tool definitions sitting in its context. Get started by installing Jentic One, the self-hosted execution layer, from its GitHub repo.
What authentication does the Flight Offers Search API use?
Amadeus uses OAuth 2.0 with the client_credentials grant. You exchange your Amadeus client_id and client_secret for an access token at the Amadeus token endpoint /v1/security/oauth2/token, sending grant_type=client_credentials in a form-encoded body, then send the token as Authorization: Bearer on each /shopping/flight-offers request. Amadeus's authorization guide shows the token response carrying expires_in of 1799 seconds. Note that this specification declares no securityScheme of its own; Amadeus documents the OAuth 2.0 flow in a separate Authorization specification and in its developer guides. Through Jentic, your client id and secret are stored encrypted by your own Jentic One instance and injected at execution time.
Can I search multi-city itineraries with the Flight Offers Search API?
Yes. Use POST /shopping/flight-offers with a structured request body containing multiple originDestinations entries. Each entry has its own origin, destination, and date, which lets you build itineraries with three or more legs in a single call. The GET endpoint only supports round-trip and one-way.
What are the rate limits for the Flight Offers Search API?
Amadeus publishes its limits per environment in the API Rate Limits developer guide at amadeus4dev.github.io/developer-guides/api-rate-limits/. Flight Offers Search falls under the general Self-Service bucket, not the artificial-intelligence and partner bucket, so the guide states 10 transactions per second per user in the test environment and 40 transactions per second per user in production, with no more than 1 request every 100ms. That guide states no monthly transaction cap. Treat 429 responses as a signal to back off and retry with a delay.
How do I search flights with the Flight Offers Search API through Jentic?
Install Jentic One using the one-line installer from its GitHub repo, run jentic register on your agent machine, then import Flight Offers Search from the API Directory and store your Amadeus client id and secret once. Your agent then searches Jentic for search flight offers, loads the operation schema, and executes either the GET or POST flight-offers operation with the user's travel parameters. Keep the agent on a separate machine from the instance, as the hardening guide describes.
Are the prices returned by the Flight Offers Search API guaranteed?
No. Prices reflect real-time airline inventory at the moment of the search and can change before booking. To confirm a price, call the Flight Offers Price endpoint with the selected offer before proceeding to booking; only the price returned by that confirmation step is bookable.
Can I limit what my agent is allowed to do with the Flight Offers Search API?
Yes. Because you run Jentic One yourself, your own rules decide which operations and credentials the agent may use, so you can grant only the GET /shopping/flight-offers search for simple round-trip queries, only the POST /shopping/flight-offers search for multi-city itineraries, or both. Neither operation books, tickets, or services a reservation, so those actions stay out of the agent's reach entirely. Your Amadeus client id and secret are held by your own instance and injected only at execution time, never exposed to the agent's prompt or context.