A new open-source project called Travel Hacking Toolkit brings AI-powered travel optimization to coding agents like OpenCode and Claude Code. The toolkit provides drop-in skills and MCP (Model Context Protocol) servers that enable users to search award flight availability across 25+ loyalty programs, compare points versus cash prices, check loyalty balances, and receive personalized travel recommendations. Created by a developer and shared on Hacker News, the project addresses a practical use case for agentic AI: helping travelers answer the fundamental question of whether to redeem points or pay cash for flights.
The toolkit includes five free MCP servers requiring no API keys—Skiplagged, Kiwi.com, Trivago, Ferryhopper, and Airbnb—alongside eight skills that connect to APIs like Seats.aero for award availability, AwardWallet for loyalty tracking, Duffel for real-time flight search, and SerpAPI for Google Flights pricing. The architecture distinguishes between MCP servers for general search tools and markdown-based "Skills" that instruct AI agents on calling loyalty program APIs. Static JSON datasets provide reference data on airline alliances, transfer partners, and points valuations not available via real-time APIs.
Community response on Hacker News highlighted both the promise and complexity of automated travel hacking. Users with substantial point balances noted challenges in finding valuable redemptions, particularly for families needing multiple seats with scheduling constraints. Commenters pointed out additional considerations including discount hotel sites like super.com and credit card portal benefits such as Amex's Fine Hotels & Resorts. The discussion underscored a practical application for AI agents: coordinating multiple APIs and data sources to solve a concrete consumer problem. The project is available on GitHub under an MIT license.