Airport On-Time Performance

API version: 1.0.4

Auth

No authentication required.

Base URL

https://test.api.amadeus.com/v1

Setup

1. No auth setup needed
2. GET /airport/predictions/on-time -- verify access

Endpoints

1 endpoints across 1 groups. See references/api-spec.lap for full details.

airport

Method	Path	Description
GET	/airport/predictions/on-time	Returns a percentage of on-time flight departures from a given airport.

Enhanced Skill Content

Question Mapping

• "What is the on-time performance prediction for JFK?" -> GET /airport/predictions/on-time
• "How likely is my flight from LAX to be on time tomorrow?" -> GET /airport/predictions/on-time
• "Which airports have the best on-time ratings for a given date?" -> GET /airport/predictions/on-time (loop over airport codes)
• "Is there a delay risk at O'Hare on 2026-03-15?" -> GET /airport/predictions/on-time
• "What's the punctuality forecast for Heathrow next Monday?" -> GET /airport/predictions/on-time
• "Compare on-time performance between SFO and SJC for a date" -> GET /airport/predictions/on-time (two calls, compare results)
• "Should I book a connection through ATL or CLT based on delay risk?" -> GET /airport/predictions/on-time (two calls, compare)
• "What on-time score does CDG have today?" -> GET /airport/predictions/on-time
• "Get delay predictions for all major US hub airports on a date" -> GET /airport/predictions/on-time (batch calls)
• "Is NRT a reliable airport for connections next Friday?" -> GET /airport/predictions/on-time
• "How does seasonal performance vary for DEN?" -> GET /airport/predictions/on-time (multiple date calls)
• "What happens if I pass an invalid airport code?" -> GET /airport/predictions/on-time (expect 400)

Response Tips

• On-time predictions: Response contains a probability/score object -- surface the percentage or category (e.g., high/medium/low) directly rather than raw nested data. Check for data.result or data.prediction keys.
• 400 errors: Typically indicate malformed airportCode (must be 3-letter IATA) or invalid date format (use YYYY-MM-DD). Parse errors[].detail for actionable messages.
• No pagination: Single-resource endpoint returns one prediction per call -- no cursor or offset handling needed.

Anomaly Flags

• Invalid IATA code: If a 400 response mentions the airport code, proactively suggest the correct 3-letter IATA code or offer to look it up.
• Date out of range: Surface when the requested date falls outside the API's supported prediction window (typically near-term only).
• Low confidence scores: If the prediction confidence or probability is unusually low, flag it so the user can factor uncertainty into decisions.
• Rate limiting (429): Surface rate limit headers (X-RateLimit-Remaining) proactively when approaching thresholds -- Amadeus test environment has strict quotas.
• Test vs production base URL: The spec uses test.api.amadeus.com -- flag if the user appears to need production data, as test responses may use synthetic data.
• Authentication failures (401/403): Amadeus requires OAuth2 bearer tokens -- surface token expiry proactively if calls start failing.

Playbook

1. Check On-Time Performance for a Single Airport

1. Confirm the 3-letter IATA airport code (e.g., JFK, LAX, LHR).
2. Determine the target date in YYYY-MM-DD format.
3. Call GET /airport/predictions/on-time?airportCode=JFK&date=2026-03-10.
4. Extract the on-time probability or category from the response.
5. Present the result with context (e.g., "JFK has an 82% on-time rating for March 10").

2. Compare Two Airports for Connection Planning

1. Identify both candidate airport codes (e.g., ATL and CLT).
2. Set the travel date.
3. Call GET /airport/predictions/on-time for each airport in parallel.
4. Compare the on-time scores side by side.
5. Recommend the airport with the higher on-time probability, noting any caveats about confidence.

3. Batch-Check Multiple Airports

1. Define a list of target airports (e.g., top 10 US hubs).
2. Set the target date.
3. Loop through the list, calling GET /airport/predictions/on-time for each.
4. Collect and sort results by on-time score, highest first.
5. Present a ranked summary table.

4. Handle and Recover from Errors

1. Make the prediction request.
2. If a 400 is returned, parse the error detail from the response body.
3. Check for common issues: invalid IATA code (must be exactly 3 uppercase letters), malformed date (must be YYYY-MM-DD), or date out of supported range.
4. Correct the parameter and retry.
5. If errors persist, verify API authentication is valid and the test environment is reachable.

Response Tips

• Check response schemas in references/api-spec.lap for field details

References

• Full spec: See references/api-spec.lap for complete endpoint details, parameter tables, and response schemas

Generated from the official API spec by LAP