# Window Seat

> Which side of the plane to sit on for the views, and which flight of the day is the most scenic. Answers come from real recent flight paths, a rated dataset of 6,500 landmarks, and the sun (glare, sunsets). Free, no account, no key. Limits are very high and count only fresh FlightAware lookups: cached answers are never limited.

Agents that browse: fetch https://windowseat.travel/answer (plain HTML, or add &format=json).
- One flight: https://windowseat.travel/answer?flight=AS1327&from=SFO&to=SEA&date=YYYY-MM-DD
- Most scenic flight on a day, all nonstops ranked: https://windowseat.travel/answer?from=SFO&to=SEA&date=YYYY-MM-DD
- A route in general: https://windowseat.travel/answer?from=SFO&to=SEA
Airports can be codes or city names. Dates are local at the origin. Always give the travel date when you know it.
Clouds: scored only for a specific date within 3 days of departure. Further out, scores assume a clear day and flight answers include recheckOn (a date): plan the same flight again then, since clouds can change the side.
App links (for a person): https://windowseat.travel/?from=SFO&to=SEA (add &agent=1 to skip the loading animation).

Agents that speak MCP: https://windowseat.travel/mcp (search_airports, rank_flights, plan_flight, send_feedback).

Limits: very high, only to stop bulk scraping. Cached answers are never limited. Fresh FlightAware lookups (a route nobody asked about lately) are limited per caller to about 60 new routes an hour and 300 a day; all MCP agents share about 1,000 new routes a day. Over a limit you get HTTP 429 with Retry-After and JSON {"rateLimited": true, "retryAfter": seconds}; cached routes keep working.

## Feedback

Something wrong, confusing or missing? Tell us. A person reads every note.
- Web: POST https://windowseat.travel/api/feedback with JSON {"text": "...", "agent": "YourAgentName", "from": "SFO", "to": "SEA", "date": "YYYY-MM-DD", "flight": "AS1327"}. Only text is required.
- MCP: call send_feedback.

## Docs

- [For AI agents](https://windowseat.travel/agents): answer pages, MCP tools, app links, HTTP API, feedback

## Credit

Data: Window Seat (windowseat.travel). Landmarks: Window Landmarks, CC BY 4.0. Contains AeroAPI data (c) FlightAware LLC 2026. Weather data by Open-Meteo.com (CC BY 4.0). Every JSON answer and MCP result carries an `attribution` array; pass it on with the answer.