Amadeus Travel Recommendations API
Browse all Amadeus APIs✓ Official Vendor SpecData EnrichmentData Transformationoauth21 EndpointsREST
For Agents
Suggest destinations for a traveller based on cities they have liked or visited, returning ranked recommendations with relevance scores for inspiration surfaces.
Use for: Suggest destinations a traveller would like based on their previous trips, Get recommended cities for someone who likes Paris and Rome, Retrieve travel recommendations for a customer profile, Find similar destinations to a city the user already booked
Not supported: Does not return prices, availability, or bookings — use for ranked destination recommendations from a seed of cities only.
The Amadeus Travel Recommendations API returns suggested destinations for a traveller based on cities they already like. Provide a seed of one or more city codes and the API returns destinations chosen by other travellers with similar preferences, complete with relevance scores. The single endpoint is designed for inspiration surfaces — homepage carousels, 'where next' modules, and AI travel concierges that need to suggest destinations rather than wait for the user to pick one.
Install Jentic One Beta
Connect the Travel Recommendations API to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Travel Recommendations 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.
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 | sh2
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 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.
Capabilities
What an agent can do with Travel Recommendations API.
Suggest destinations from a seed list of cityCodes the traveller already likes
Return ranked recommendations with a relevance score per suggestion
Filter recommendations by traveller country code for region-aware suggestions
Power 'where to go next' inspiration surfaces in travel apps
Feed AI travel concierges with grounded destination ideas rather than hallucinated picks
Use Cases
Patterns agents use Travel Recommendations API for, with concrete tasks.
★ Personalised inspiration carousel
Power a 'destinations you might love' carousel on a travel app's homepage by feeding Travel Recommendations the city codes of the user's recent searches or bookings. The API returns ranked suggestions with relevance scores so the carousel can prioritise the strongest recommendations and refresh them as the user's history grows. Implementation is a single GET call per user session.
Call GET /reference-data/recommended-locations with cityCodes=PAR,ROM and return the top 6 recommended destinations with relevance scores.
Destination cross-sell after booking
After a user books a hotel or flight, surface recommended next destinations using the booking's city as the seed. This converts a single transaction into a future-trip nurture — useful for OTAs and corporate travel platforms that want to extend the user lifecycle. Combine with email or push notifications timed for a few weeks after the booked trip ends.
Given the user just booked a stay in BCN, call Travel Recommendations with cityCodes=BCN and return the top 3 destinations to suggest in a follow-up email.
AI travel agent destination picker
When an AI travel agent is asked 'where should I go for a long weekend' with no specific destination, Travel Recommendations grounds the answer in real similarity data rather than the model's prior. The agent supplies the user's previously enjoyed cities as the seed, retrieves ranked suggestions, and presents them with a citation of the relevance score. This avoids hallucinated city picks and lets the user confirm before chaining into Hotel Search.
Use Jentic to search 'recommend a destination', execute Travel Recommendations with the user's preferred city codes, and return the top 3 suggestions with relevance scores.
Key Endpoints
1 endpoints — the amadeus travel recommendations api returns suggested destinations for a traveller based on cities they already like.
METHOD
PATH
DESCRIPTION
GET
/reference-data/recommended-locations
Get recommended destinations for a seed of cities
GET
/reference-data/recommended-locations
Get recommended destinations for a seed of cities
Why Jentic?
Three things that make agents converge on Jentic-routed access.
Credential isolation
Amadeus OAuth2 client_id and client_secret are stored encrypted in the Jentic vault. Agents receive scoped bearer tokens; the client secret never enters the agent's context.
Intent-based discovery
Agents search by intent ('recommend a destination') and Jentic returns the Travel Recommendations operation with its cityCodes, travelerCountryCode, and destinationCountryCodes parameter schemas.
Time to first call
Direct Amadeus integration: 1 day for OAuth and ranked-list rendering. Through Jentic: under 30 minutes — search, load, execute.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Travel Recommendations API through Jentic.
What authentication does the Travel Recommendations API use?
OAuth 2.0 client credentials. Exchange your Amadeus API key and secret for a bearer access token via the authorization endpoint, then pass it as the Authorization header on each call. Through Jentic, the client secret stays encrypted in the vault and only a scoped token is exposed to the agent.
What inputs does the recommendation endpoint accept?
GET /reference-data/recommended-locations accepts cityCodes (a comma-separated list of IATA city codes the traveller likes) and optionally travelerCountryCode and destinationCountryCodes filters. The response is a ranked list of destinations with a relevance score for each.
Can I bias recommendations to a region?
Yes. Use the destinationCountryCodes parameter to scope suggestions to one or more ISO country codes — useful when the traveller has a constraint like 'within Europe' or 'domestic only'.
What are the rate limits for the Travel Recommendations API?
Amadeus enforces per-second and per-month transaction quotas that vary by environment. The test environment (test.api.amadeus.com/v1) has low quotas for development; production limits are set in your Amadeus for Developers contract.
How do I recommend a destination through Jentic?
Install with pip install jentic, then search 'recommend a destination' to discover the Travel Recommendations operation. Load its schema and execute with cityCodes seeded from the user's history. Jentic returns parsed ranked suggestions ready to render. Get started at https://app.jentic.com/sign-up.
Is the Travel Recommendations API free?
Amadeus offers a free self-service tier suitable for prototyping. Production usage requires upgrading to a paid plan in the Amadeus for Developers portal; pricing depends on call volume and contract.
GET STARTED