Airtreks MCP Server

AirTreks MCP Server

The complex-itinerary tool for AI agents. Multi-stop, round-the-world, open-jaw, surface segments - the trips that standard flight search can't handle. When your user asks for 3+ stops across continents, this server answers with routing analysis built on real AirTreks fare-construction data: 60+ carriers, 53 known dead legs, bookability rates measured from 1,400+ real fare attempts, and 20 proven routing templates from actual bookings.

Live endpoint: https://mcp.airtreks.com/mcp - free, no API key, 100 requests/day.

Now with real prices: a historical range for any route in under a second, live fares for a specific itinerary, and a full multi-stop trip priced across several tickets — the same trip can vary three-fold depending on how it is ticketed, and that is the part no other flight tool will tell you.

30-second setup

Claude Desktop

Settings → Connectors → Add custom connector → paste https://mcp.airtreks.com/mcp.

Claude Code

claude mcp add --transport http airtreks https://mcp.airtreks.com/mcp

Cursor

Add to .cursor/mcp.json (or Cursor Settings → MCP → Add new server):

{
  "mcpServers": {
    "airtreks": {
      "url": "https://mcp.airtreks.com/mcp"
    }
  }
}

No API key is needed for any tool, the consultant handoff included. A key is optional and only raises your daily limit; append it to the URL if your client cannot send custom headers, as claude.ai connectors cannot:

https://mcp.airtreks.com/mcp?key=at_your_key

ChatGPT

Settings → Apps & Connectors → enable Developer mode (Pro/Business plans) → add https://mcp.airtreks.com/mcp as a custom connector.

Any other MCP client

Same JSON shape as Cursor above (Streamable HTTP). Prefer stdio? Run it locally:

npx airtreks-mcp

Then ask your agent:

Plan a round-the-world trip: San Francisco, Tokyo, Bangkok, Singapore, Delhi, Istanbul, London, back to San Francisco.

One call, one answer: a 7-leg RTW

Your user wants six stops around the world. Google Flights gives up. One plan_route call lays the trip out, and one route_estimate call says what it costs. Neither one picks airlines: AirTreks consultants do that when they build the ticket.

{
  "cities": ["SFO", "NRT", "BKK", "SIN", "DEL", "IST", "LHR", "SFO"],
  "budget": "mid"
}

Real plan_route response, trimmed for length (the full version has notes for all 7 legs):

{
  "route": "SFO -> NRT -> BKK -> SIN -> DEL -> IST -> LHR -> SFO",
  "totalLegs": 7,
  "isRoundTrip": true,
  "direction": "westbound",
  "backtracking": false,
  "regionsCrossed": ["americas", "asia", "europe"],
  "customBuild": {
    "strategy": "7-leg route. 1 surface sector opportunity.",
    "surfaceSectors": [
      {
        "insteadOf": "BKK -> SIN (leg 3)",
        "suggestion": "Bangkok to Singapore through Malaysia. Train, bus, or ultra-cheap LCC. Adds Malaysia and possibly Penang, KL, Melaka.",
        "savings": "Saves $100-250"
      }
    ],
    "complexity": "moderate"
  }
}

And route_estimate on the same cities:

{
  "cabin": "economy",
  "currency": "USD",
  "estimate": { "low": 1616.43, "high": 2465.75 },
  "confidence": "medium",
  "basis": "1,616-2,466 USD per person, from AirTreks fare history. We have history for most of this route, but not all of it."
}

That is a range from past AirTreks bookings, not a quote. For a real number, trip_idea_create hands the complete trip - route, a date for every leg, passengers, cabin and the traveler's answers to our planning questions - to an AirTreks consultant who prices and books it.

Tools

All tools are free and need no API key — 100 requests/day per IP.

Routing

Pricing

Consultant handoff

Submits real customer contact details, so only call it with the customer's knowledge and consent. Repeat submissions of the same email + route inside 24 hours return the existing trip request rather than creating a duplicate.

Why this data is different

AirTreks has built complex multi-stop itineraries since 1987. This server exposes what that history taught us:

• Dead legs - city pairs that look bookable but fail on alliance fares, learned from 1,400+ real fare-construction failures
• Bookability rates by leg count - measured, not estimated (91% at 3-4 legs, 61% at 5-6, 6% at 7+)
• Poison carriers and hub fixes - we know what airline combinations break ticketing and what to route instead
• Proven routing templates from working with Multi-Stop and Around the World Intineraries for inviduals, families and large group

No other flight tool returns this because no other flight tool has priced these failures.

REST API (no MCP client required)

Every tool is also a plain REST endpoint - same tools, same rate limits, JSON in/out. For agent frameworks that consume REST instead of MCP:

curl -X POST https://mcp.airtreks.com/api/plan_route \
  -H "Content-Type: application/json" \
  -d '{"cities": ["LAX", "NRT", "BKK", "LHR", "LAX"]}'

Full OpenAPI 3.1 spec: https://mcp.airtreks.com/openapi.json

Rate limits

• Free: 100 requests/day per IP, no key needed — every tool, trip_idea_create included
• Registered: higher limits with an API key (X-API-Key header, or ?key= for clients that cannot set headers)

Endpoints

License

AGPL-3.0-only