AI agent skill for world-class travel planning using Kontour AI's 9-dimension progressive planning model.
npx skills add Bookingdesk-AI/kontour-travel-plannerOr browse on skills.sh and clawhub.com.
Transforms any AI agent into a travel planning consultant using a structured methodology:
- 9 weighted dimensions — dates, destination, budget, duration, travelers, interests, accommodation, transport, constraints
- 4-stage conversation flow — Discover → Develop → Refine → Confirm
- Guided discovery — one high-impact question per turn, concrete options, conflict detection
- Structured output — trip context JSON, day-by-day itinerary, budget breakdown, Google Maps export
- Reference data — 200 destinations, 500 airports, airlines, activities, budget benchmarks (no API needed)
Ground truth files in references/:
| File | Contents |
|---|---|
destinations.json |
200 global destinations with coordinates, costs, best months |
airports.json |
500 airports with IATA codes and coordinates |
airlines.json |
Major airlines with alliances, hubs, regions |
activities.json |
Activity types with durations, cost tiers |
budget-benchmarks.json |
Daily cost benchmarks by destination tier |
booking-integrations.json |
Integration roadmap for booking providers |
embed-snippets.json |
Embeddable widget templates |
scripts/plan.sh— Get structured trip context from natural languagescripts/export-gmaps.sh— Export itinerary to Google Maps links and KMLscripts/gen-airports.py— Generate airport reference data
- Kontour AI — Interactive map planning and booking
- GitHub
MIT