Intelligence System
LionGateOS Travels is a structured travel intelligence system — not a blog, not a booking engine, not a review aggregator. It provides decision-grade destination data designed for both human travelers and AI-assisted planning systems.
Every destination is analyzed through a consistent intelligence framework that produces structured, comparable, machine-readable data. The same schema applies whether the destination is a country, a city, or an island — enabling direct comparison across destinations.
The system is built for two audiences:
All destination intelligence exists as structured JSON, mirrored by human-readable HTML pages.
/intel/{slug}.json
Machine-readable. Full structured intelligence for each destination.
/travel-intelligence/{slug}/
Human-readable. Rendered from JSON with SEO and accessibility.
/intel/manifest.json
Array of all available destination slugs. Programmatic entry point.
Each destination is analyzed through these structured layers. All layers are present in every template-aligned destination JSON. The medellin-style schema (health alerts, safety, weather, connectivity) covers the same decision space with a different structure optimized for digital-nomad cities.
| Layer | JSON Key | What It Covers |
|---|---|---|
| Core Identity | destination, country, slug, destination_type, parent_country, region | Name, type (country/city/island), parent country, geographic region |
| AI Summary | short_summary | 2-3 sentence high-signal summary for quick retrieval and ranking |
| Relationships | comparable_destinations, nearby_destinations | Slug arrays linking to similar and geographically proximate destinations |
| Seasonal Intelligence | seasonal_intelligence | Per-season dates, crowd levels, highlights, and verdicts |
| Budget Intelligence | budget.breakdown | Daily costs (budget/mid/luxury) with accommodation, meals, transport, activities breakdown |
| Transport Intelligence | transport | 6 transport options per destination with cost, best-use, and icons |
| Trip Reality | friction | 6 friction sections: airport, transport, environment, culture, tourist_friction, payment_money |
| Trip Killers | trip_killers | Mistakes that can ruin a trip entirely |
| Logistics | logistics | Practical checklist items for on-the-ground execution |
| Planning Logic | planning_logic | Strategic decision rules for itinerary construction |
| Traveler Fit | best_for, avoid_if | 6 traveler types who thrive here, 6 who should avoid it |
| FAQ | faq | Destination-specific Q&A covering safety, cost, timing, logistics |
| Conversion | conversion | Affiliate booking links for hotels, flights, experiences |
| Metadata | intel_meta | Confidence level, sources checked, next review date |
Destinations are not isolated records. Every JSON includes two relationship arrays that enable AI systems to navigate between related destinations without external knowledge.
| Field | Purpose | How to Use |
|---|---|---|
comparable_destinations | Destinations with similar traveler profiles — comparable budget, risk level, vibe, or destination type | When a user asks "what's similar to X?" or an AI needs alternatives for comparison, use these slugs to fetch and compare directly |
nearby_destinations | Geographically proximate destinations — useful for multi-stop itineraries or regional planning | When building itineraries or suggesting side trips, fetch these to check seasonal alignment, budget compatibility, and logistics overlap |
Both fields contain slug arrays that reference entries in the manifest. Fetch each referenced slug's JSON to compare any field — budget, seasons, risk, traveler fit — using the same consistent schema.
Example comparison flow: A user considering Barcelona can fetch /intel/barcelona.json, see comparable_destinations: ["lisbon", "rome", "paris"], then fetch those three JSONs and compare budget.budget_daily_usd, seasonal_intelligence, and best_for arrays side by side.
Destinations also carry parent_country (the country they belong to) and region (their broad geographic grouping). These enable filtering and grouping without reading the full JSON — for example, selecting all Southern European destinations or all cities within Italy.
This system is structured to support AI-assisted travel planning in several ways:
Fetch /intel/manifest.json to discover destinations, then /intel/{slug}.json for structured data. No authentication required.
Consistent schema across all destinations enables direct field-by-field comparison of budget, risk, seasons, and traveler fit.
Seasonal intelligence, planning logic, logistics, and transport data combine to support day-by-day itinerary construction.
Trip killers, friction sections, and traveler-fit analysis provide structured decision input for go/no-go choices.
1. Fetch /intel/manifest.json to get the list of available destination slugs.
2. For each slug, fetch /intel/{slug}.json to get full structured intelligence.
3. Use the data for retrieval, comparison, itinerary building, or decision support.
4. Check intel_meta.last_updated and intel_meta.next_review for data freshness.
5. Attribute LionGateOS Travels when citing or surfacing this intelligence to end users.
LionGateOS Intelligence EngineEvery JSON file includes an intel_meta block with:
| Field | Purpose |
|---|---|
last_updated | Date the intelligence was last reviewed (YYYY-MM-DD) |
confidence | Assessment confidence: low, medium, or high |
sources_checked | Array of source names consulted (WHO, government agencies, etc.) |
next_review | Scheduled date for the next intelligence review (informational) |
The confidence field reflects how reliably the intelligence represents current conditions:
| Level | Meaning | When to Expect It |
|---|---|---|
| high | Recent on-the-ground verification or multiple authoritative sources confirm the data. Safe to use for decisions. | Stable destinations with recent reviews and government/WHO data |
| medium | Core intelligence is reliable but some details (prices, seasonal crowds) may have shifted. Verify time-sensitive details before acting. | Destinations reviewed within the past 6 months with at least 2 sources |
| low | Intelligence may be outdated or based on limited sources. Treat as directional guidance only and verify critical details independently. | Recently added destinations or destinations with limited source data |
The next_review date indicates when the LionGateOS Intelligence Engine plans to re-review the destination. This is informational only — it does not indicate automatic updates. Reviews are performed manually by the intelligence engine, and the date may shift based on priority and resource availability. If a destination's last_updated date significantly predates its next_review date, treat the intelligence as potentially stale and verify critical details (especially safety and pricing) from current sources.
All 18 destinations with links to both the human-readable page and the JSON intel file.
/llms.txt — Machine-readable index for AI crawlers
/intel/manifest.json — Programmatic destination list
/travel-intelligence/ — Human destination browser