Tool description (what the LLM sees)
Tool description (what the LLM sees)
Find the best date options for a known origin and destination using cached pricing. Takes the SAME inputs as flight_calendar, but returns up to ten cheapest itineraries, one per (departure, return) date-pair, spread across the requested departure window, so the user sees a diverse set of "when should I fly?" options rather than the full per-date list.
WHEN TO USE THIS TOOL:
- The user knows the route (origin AND destination) and is flexible on WHEN, and wants a short, spread-out shortlist of the best date options.
- Examples: "When is the cheapest time to fly Paris → New York in June?", "Best weekends to go from London to Lisbon next month", "Give me a few good date options for a week in Tokyo".
WHEN TO USE flight_calendar INSTEAD:
- The user wants the full price-per-date list / calendar grid for a route, not a curated shortlist.
WHEN TO USE flight_search INSTEAD:
- The user has committed to EXACT dates (single departure AND single return). flight_search hits live pricing and is the step before booking.
IMPORTANT:
All dates MUST be in the future. Never use past dates.
Origin and destination must be IATA city codes by default, or IATA airport codes if the user specifies them.
By default search roundtrip unless the user explicitly asks for one-way (trip_type="oneway").
Flow: find_dates → flight_search (price-check the chosen dates) → trip → checkout
PRICES: in the text output, price is already a decimal in major units, display as-is, do NOT divide. In structuredContent, flights[].total_amount and origin_total_amount are objects {value, currency, decimal_places} where value is an INTEGER in minor units, divide value by 10^decimal_places before displaying those (value 15977 with decimal_places 2 is 159.77 USD).
Parameters
| Name | Type | Required | Description |
|---|---|---|---|
origin | string | Yes | Origin airport IATA code or city code (3 letters). Examples: “JFK”, “PAR”. |
destination | string | Yes | Destination airport IATA code or city code (3 letters). Examples: “CDG”, “NRT”. |
trip_type | enum ("oneway" | "roundtrip") | Yes | Trip type: “oneway” or “roundtrip”. |
departure_dates | array<string> | No | Specific departure dates (YYYY-MM-DD). Multiple dates searched with OR logic. |
departure_date_ranges | array<object> | No | Departure date ranges for flexible search. |
departure_date_ranges[].start | string | Yes | Range start date (YYYY-MM-DD) |
departure_date_ranges[].end | string | Yes | Range end date (YYYY-MM-DD) |
return_dates | array<unknown> | No | Specific return dates (YYYY-MM-DD) for round-trip flights. |
return_date_ranges | array<object> | No | Return date ranges for flexible round-trip search. |
return_date_ranges[].start | string | Yes | Range start date (YYYY-MM-DD) |
return_date_ranges[].end | string | Yes | Range end date (YYYY-MM-DD) |
stay_days | integer | No | Exact stay duration in days. Auto-calculates return date. |
stay_days_range | object | No | Flexible stay duration range. Example: {min: 5, max: 10} |
stay_days_range.min | integer | Yes | Minimum stay duration in days |
stay_days_range.max | integer | Yes | Maximum stay duration in days |
direct_only | boolean | No | Only return nonstop flights. |
cabin_class | enum ("economy" | "premium_economy" | "business" | "first") | No | Cabin class filter. |
max_total | number | No | Maximum total price for the trip (canonical name). Preferred over max_price. |
sort_by | enum ("lowest" | "recommendation") | No | Sort order. Default: “lowest” (cheapest first). |
currency | string | No | ISO 4217 currency code. Defaults to “USD”. |
locale | string | No | BCP 47 locale. Defaults to “en-US”. |
format | enum ("text" | "json") | No | Response format. “text” returns plain text for LLMs. “json” returns structured JSON. |
limit | integer | No | Max date-pairs to return (1-100). Default 20. |
user_intent | string | No | The end user’s current request in their own words, e.g. “find a cheap flight to Tokyo in mid-June”. Pass it as-is when short; otherwise condense the goal and constraints into 1-2 sentences. Strip personal identifiers (names, emails, phone numbers, addresses, payment details), replace them with placeholders like “[name]”. Optional and never changes the result of the call; Jinko uses it to understand demand and improve results. |