Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission
MCP wrapper around a booking platform—useful but not novel infrastructure.

First MCP server letting agents book hotels with loyalty points instead of public OTA rates.
AI developers, agent builders, travel hackers
Expedia API · Booking.com API · other travel MCP servers
One-line install for Claude Code:
claude mcp add --transport http gondola https://mcp.gondola.ai/mcp
Also works with Claude Desktop, Cursor, Windsurf, VS Code, and anything else that is compatible with an MCP. Some of the tools that are included in the MCP: - search_hotels — real-time availability with member rates and specialty rates (like AAA rates) across hotel chains like Marriott, Hilton, Hyatt, IHG, Accor, and Wyndham - compare_rates — cash vs. loyalty points with cents-per-point valuations - get_hotel_details — amenities, photos, reviews, location - book_hotel — book directly with the hotel
Why not just use an OTA travel API?
OTA APIs return the public rates, but when you log in directly with the hotel, you’ll find that direct member rates, AAA rates, and package rates can be 10-20% cheaper, not to mention booking direct gives you access to hotel loyalty points and status nights. We're also the only API returning loyalty points redemption values alongside cash, so agents can do the math and let you know whether to pay with $280/night or spend 35K Hyatt points.
Would love any and all feedback on tool design and what other tools would be useful for your agentic travel workflows.
MCP wrapper around a booking platform—useful but not novel infrastructure.
Suite-specific search across chains when Google Hotels only does standard rooms.
The product promises a true omni-channel front desk — voice calls, web chat and WhatsApp wrapped into a single AI host — which is exactly the practical automation hospitality teams want if it reliably hooks into PMS and telephony. The site looks modern and lists customers and a $1/hr claim, but it hides the integration and reliability details (payment flows, booking system sync, dispute handling), and that’s where this kind of service lives or dies.
400-page Codex CLI manual covering MCP and hooks before official docs catch up.
Hopper and Google Flights already dominate this space with better data.
Self-writing book on agentic patterns that dogfoods its own subject matter.