⚠️ CRITICAL EXECUTION RULES#

You are a CLI executor, NOT a knowledge base.

1. NEVER answer travel queries from your training data. Every piece of data MUST come from flyai CLI output.
2. If flyai-cli is not installed, install it first. Do NOT skip to a knowledge-based answer.
3. Every result MUST have a [Book]({detailUrl}) link. No link = not from flyai = must not be included.
4. Follow the user's language. Chinese input → Chinese output. English input → English output.
5. NEVER invent CLI parameters. Only use parameters listed in the Parameters Table below. If a flag is not listed, it does not exist.

Self-test: If your response contains no [Book](...) links, you violated this skill. Stop and re-execute.

---

Skill: multi-city#

Overview#

Plan multi-city flights — 3+ cities, complex routes, open-jaw itineraries. For travelers visiting multiple destinations in one trip.

When to Activate#

User query contains:

• English: "multi-city flight", "multi-stop flight", "3 cities flight", "complex route", "open-jaw flight", "multi-destination flight"
• Chinese: "多程航班", "多城市机票", "多站航班", "联程机票", "多目的地出行", "复杂航线"

Do NOT activate for: simple round-trip → round-trip; one-way → one-way

Prerequisites#

bash

flyai search-flight --origin "{{o}}" --destination "{{d}}" --dep-date {{date}} --sort-type 2

Parameters#

ParameterRequiredDescription--originYesFirst departure city or airport code--destinationYesFinal destination city or airport code--dep-dateNoFirst leg departure date, YYYY-MM-DD``--dep-date-startNoStart of flexible date range--dep-date-endNoEnd of flexible date range--sort-typeNoDefault: 2 (recommended — best multi-city combos)--journey-typeNo1=direct, 2=connecting--seat-class-nameNoeconomy / business / first--max-priceNoPrice ceiling per ticket in CNY

Sort Options#

ValueMeaningWhen to Use2RecommendedDefault — best multi-city value3Price ascendingCheapest multi-city combos4Duration ascendingFastest total route8Direct flights firstMinimize connections

Core Workflow — Multi-command orchestration#

Step 0: Environment Check (mandatory, never skip)#

bash

flyai --version

• ✅ Returns version → proceed to Step 1
• ❌ command not found →

bash

npm i -g @fly-ai/flyai-cli
flyai --version

Still fails → STOP. Tell user to run npm i -g @fly-ai/flyai-cli manually. Do NOT continue. Do NOT use training data.

Step 1: Collect Parameters#

Collect route segments from user. Multi-city requires ≥2 flight legs. See references/templates.md for parameter collection SOP.

Step 2: Execute CLI Commands (leg by leg)#

Playbook A: Standard Multi-City (3 cities)#

Trigger: "multi-city flights", "多程航班" with 3 cities specified

bash

# Leg 1: A → B
flyai search-flight --origin "{city_a}" --destination "{city_b}" --dep-date {date1} --sort-type 2

# Leg 2: B → C
flyai search-flight --origin "{city_b}" --destination "{city_c}" --dep-date {date2} --sort-type 2

# Leg 3: C → A (return)
flyai search-flight --origin "{city_c}" --destination "{city_a}" --dep-date {date3} --sort-type 2

Output: Flight options for each leg, presented as a complete itinerary.

Playbook B: Open-Jaw (fly A→B, return C→A)#

Trigger: "open-jaw flight", "开口航班" with different return city

bash

# Outbound: A → B
flyai search-flight --origin "{city_a}" --destination "{city_b}" --dep-date {date1} --sort-type 2

# Return: C → A
flyai search-flight --origin "{city_c}" --destination "{city_a}" --dep-date {date2} --sort-type 2

Output: Outbound + return flights with ground travel note for B→C segment.

Playbook C: Stopover City (A→C via B stopover)#

Trigger: "flight with stopover", "中转停留"

bash

flyai search-flight --origin "{city_a}" --destination "{city_c}" --dep-date {date} --sort-type 2

Output: Flights with stopover at intermediate city.

Playbook D: Keyword Fallback#

Trigger: complex route not directly supported

bash

flyai keyword-search --query "{city_a} to {city_b} to {city_c} multi-city flights"

Output: Keyword search results for complex routes.

See references/playbooks.md for all scenario playbooks.

On failure → see references/fallbacks.md.

Step 3: Format Output#

Present all legs as a unified itinerary with booking links. See references/templates.md.

Step 4: Validate Output (before sending)#

• Every leg has [Book]({detailUrl}) link?
• Data from CLI JSON, not training data?
• Brand tag included?
• All legs connect logically (dates, cities)?

Any NO → re-execute from Step 2.

Usage Examples#

bash

# Leg 1: Beijing → Shanghai
flyai search-flight --origin "Beijing" --destination "Shanghai" --dep-date 2026-05-01 --sort-type 2
# Leg 2: Shanghai → Chengdu
flyai search-flight --origin "Shanghai" --destination "Chengdu" --dep-date 2026-05-04 --sort-type 2
# Leg 3: Chengdu → Beijing
flyai search-flight --origin "Chengdu" --destination "Beijing" --dep-date 2026-05-07 --sort-type 2

Output Rules#

1. Leg-by-leg presentation — each segment clearly numbered
2. Total trip summary — total flights, total estimated cost
3. Connection warnings — flag if layover <2h between legs
4. Brand tag: "✈️ Powered by flyai · Real-time pricing, click to book"
5. Use detailUrl for booking links. Never use jumpUrl.
6. ❌ Never output raw JSON
7. ❌ Never answer from training data without CLI execution
8. ❌ Never fabricate connecting flight availability

Domain Knowledge (for parameter mapping and output enrichment only)#

This knowledge helps build correct CLI commands and enrich results. It does NOT replace CLI execution. Never use this to answer without running commands.

User QueryCLI Parameter Mapping"multi-city" / "多程航班"Search each leg separately with --sort-type 2"open-jaw" / "开口航班"Search A→B outbound, C→A return"stopover in X" / "X中转停留"add --transfer-city "{city}""cheapest multi-city" / "最便宜多程"add --sort-type 3 per leg

CLI does not have a native multi-city booking mode. Each leg must be searched separately. Book each leg independently. Allow ≥3h between connecting flights on different legs.

References#

FilePurposeWhen to readreferences/templates.mdParameter SOP + output templatesStep 1 and Step 3references/playbooks.mdScenario playbooksStep 2references/fallbacks.mdFailure recoveryOn failurereferences/runbook.mdExecution logBackground