This is Layer 4 execution infrastructure for ground transfers in Latin America, currently focused on Quito, Ecuador. It exposes five partner-only tools: safe pickup point lookup, operational time windows with traffic buffers, ANT carrier permit validation, zone safety classification, and hotel operational context. All data comes from a 17-zone Place Graph with sub-50ms latency and driver-verified safety scores. You'd reach for this if you're building travel AI that needs to bridge the gap between a booking confirmation and an actual verified pickup, especially when ANT compliance and real-time safety intel matter. Currently in soft launch with NH Collection Quito through Q2 2026.
Ground truth operational layer for urban transfers in Latin America.
https://oracle.whaber.ai/.well-known/agent-card.json
Whaber Oracle is the Layer 4 execution infrastructure for travel AI in LatAm — the layer between a booking and a verified ground transfer.
Any A2A-compatible agent can discover and consume the Oracle via the canonical endpoint above.
| Tool | Description |
|---|---|
| whaber_get_safe_pickup_point | Physically verified pickup point for a hotel or airport in Quito |
| whaber_calculate_ops_window | Operational time window with Quito traffic buffers and ANT compliance |
| whaber_validate_carrier_compliance | ANT operating permit validation for Ecuador carriers |
| whaber_query_place_safety | Zone safety classification (safe / caution / avoid) for a time window |
| whaber_get_hotel_context | Operational context for hotels registered in the Place Graph |
NH Collection Quito. Partner-Only soft launch Q2 2026.
Whaber Travel-Ops · oracle@whaber.ai · https://whaber.ai
io.github.ericm1018/skillfm-llm-cost-optimizer-openai-anthropic-usage
io.github.mikerawsonnz/llm-orchestration-agent
io.github.mikerawsonnz/authenticated-llm-agent
labforgedev/copilot-memory-mcp
csoai-org/agent-prompt-injection-firewall-mcp
io.github.mikerawsonnz/authenticated-multi-llm-agent