Flight Agent Matrix icon Travel MCP

Flight Agent Matrix MCP

Compare flight options across Google Flights, Skyscanner, Booking.com, and Skiplagged with one MCP call for route, dates, class, passenger counts, stop limits, and duration caps.

  • check_circle One public tool for one-way or roundtrip provider comparison
  • check_circle Structured output with ranked cards and provider booking links
  • check_circle Rich UX with filtering and sorting when used with ChatGPT web
  • check_circle Use in Codex to automate your search and monitor prices
Flight Agent Matrix MCP demo showing the compare_flights_all_providers MCP call and widget output

Works with ChatGPT

Flight Agent Matrix rich ChatGPT experience with filtering and sorting

Rich UX allows you to quickly filter and sort results

Automate your search with Codex Agents

Learn more at our github

Codex Agents automating Flight Agent Matrix searches

FAQ

Common questions teams ask before integrating Flight Agent Matrix MCP.

How do I install this?

It's as simple as setting up your Infoseek account to get your app URL. You can paste that URL into your preferred AI.

What does Flight Agent Matrix MCP do?

It compares flights across Google Flights, Skyscanner, Booking.com, and Skiplagged with one MCP call and returns ranked, structured output for both reasoning and UI rendering.

Can I use it in ChatGPT and Codex?

Yes. The response supports rich filtering and sorting in ChatGPT and can be used by Codex Agents to automate repeated travel searches and price monitoring.

What inputs are required?

Required fields are from_airports, to_airports, and departure_date. Optional fields include return_date, max_stops_allowed, max_total_duration_hours, flight_class, adults, and children.

Are search results real time?

Searches are live at request time. Pricing and availability can change quickly, so run a fresh query before booking.

What providers are searched?

Flight Agent Matrix compares Google Flights, Skyscanner, Booking.com, and Skiplagged in one tool call.

How do I automate searches and get a daily e-mail?

Run the MCP from a scheduled Codex Agent or workflow and send the ranked results to email using your preferred mail step.

Will you provide a skill that I can use in Codex?

Yes. We can provide a Codex skill pattern that wraps the MCP search call and supports recurring monitoring flows.

How long do searches take?

Most searches return in seconds, depending on route complexity, provider response times, and filter constraints.

Can I search one-way and roundtrip flights?

Yes. One-way uses departure_date only. Roundtrip uses both departure_date and return_date in the same call.

Can I filter by stops and total duration?

Yes. Use max_stops_allowed and max_total_duration_hours to constrain itineraries before ranking.

Does it return booking links?

Yes. Top ranked options include provider booking URLs so users can open and complete booking directly.

Can I use this in my own MCP client?

Yes. Any MCP client that supports tools/list and tools/call can integrate this endpoint.

Can teams use this for travel operations?

Yes. Teams use it for itinerary comparison, route monitoring, and recurring travel-search automations.

Tools

The public MCP surface is intentionally simple: one comparison tool handles multi-provider itinerary search and returns ranked output that works for both reasoning and UI rendering.

Tool

compare_flights_all_providers

flight

Use this tool to compare flights across providers for an exact route and date set. Send one full itinerary per call. For roundtrip searches, include both departure_date and return_date in the same request.

Required inputs

  • from_airports
  • to_airports
  • departure_date

Useful optional inputs

  • return_date
  • max_total_duration_hours
  • max_stops_allowed
  • flight_class
  • adults, children

Returns

  • results[]
  • count
  • search
  • cards[]

Response JSON

The MCP returns both ranked raw flight groups and a pre-shaped cards array for immediate UI rendering. That gives agents one payload for reasoning and presentation.

cards[]

Best path for UI rendering. Each card includes price, route labels, outbound and return summaries, and booking URLs.

results[]

Structured ranked groups for custom agent-side ranking, filtering, and follow-up transformations.

search

Compact route and date summary that is useful for cards, logs, and fallback text.

count

Quick result cardinality without scanning the full response array.

{
  "count": 25,
  "search": {
    "route_label": "SEA → LAX",
    "date_label": "Apr 11 - Apr 15",
    "trip_type": "round_trip"
  },
  "cards": [
    {
      "id": "flight-card-1",
      "badge": "Best value",
      "price_display": "$325",
      "booking_url": "https://www.google.com/travel/flights/...",
      "outbound": {
        "airline": "Alaska Airlines",
        "duration": "2h 45m",
        "stops": 0
      }
    }
  ],
  "results": [
    {
      "flight_result_group": {
        "rank": 1,
        "round_trip_price": "$325",
        "google_flights_link": "https://www.google.com/travel/flights/..."
      }
    }
  ]
}

Quickstart + Code

The MCP uses standard JSON-RPC methods. The public surface includes tools/list, tools/call, resources/list, and resources/read. The Flight Agent Matrix widget is exposed at ui://widgets/flight-agent-matrix.

POST /mcp/flight-agent-matrix
{
  "jsonrpc": "2.0",
  "id": "1",
  "method": "tools/call",
  "params": {
    "name": "compare_flights_all_providers",
    "arguments": {
      "from_airports": ["SEA"],
      "to_airports": ["SFO"],
      "departure_date": "2026-10-24",
      "return_date": "2026-10-30",
      "max_total_duration_hours": 8,
      "max_stops_allowed": 0,
      "flight_class": "economy",
      "adults": 1
    }
  }
}

Access + Auth

This MCP exposes a standard MCP transport surface. In practice, teams usually consume it through Infoseek routing or a direct MCP client that speaks JSON-RPC and can call tools/list and tools/call.

The widget contract is built in: the tool advertises openai/outputTemplate and returns card-shaped output that can render directly into chat-based AI interfaces.

Why These Outputs Matter

  • Better agent ranking: results[] retains ranked itinerary groups for reasoning layers.
  • Faster UI rendering: cards[] is already shaped for chat widgets and side panels.
  • Better link persistence: top options carry provider booking URLs you can store or re-open later.

Use Cases

Flight comparison agents

Run one MCP call to compare Google Flights, Skyscanner, Booking.com, and Skiplagged results in parallel.

Travel planning copilots

Add stop and duration constraints directly into the tool arguments instead of trying to post-filter fuzzy search output later.

Widget-first chat apps

Use the built-in output template and pre-shaped cards to drop multi-provider results straight into an MCP-enabled UI.

Need this MCP in your stack?