air_choose
MCP server that watches flight and train fares across multiple provider APIs, with tools to manage watched routes and a background scheduler that re-checks hourly for new offers.
README
<p align="center"> <h1 align="center">Air Choose</h1> <p align="center"> <strong>MCP server for monitoring flight prices across multiple airline providers</strong> </p> </p>
<p align="center"> <img src="https://img.shields.io/badge/Node.js-20+-339933?logo=node.js&logoColor=white" alt="Node.js"> <img src="https://img.shields.io/badge/MCP-1.29+-365DFF?logo=modelcontextprotocol&logoColor=white" alt="MCP"> <img src="https://img.shields.io/badge/license-MIT-yellow" alt="License"> </p>
Read-only MCP server that monitors flight prices across multiple airline providers. An AI agent (Hermes, OpenClaw, Claude, etc.) adds, removes, and inspects watched routes through MCP tools; a background scheduler re-checks every watch hourly and queues new/cheaper offers for the agent to pick up. No ticket ordering — only available flight data.
Features
- Multi-provider — Kupibilet (worldwide), Ryanair (Europe), SpiceJet (India), Jetstar Asia (Asia-Pacific)
- Flights only — trains and other transport types are filtered out automatically
- Smart resolution — turn
"Moscow → Thailand"into provider-specific airport/country codes - Price drop detection — tracks best-ever prices per route, queues notifications when fares drop
- MCP-native — works with any MCP-compatible agent host
Quick Start
Local
git clone https://github.com/yourusername/air-choose.git
cd air-choose
npm install
node src/server.js
Docker
docker compose up -d
Wire into your agent
{
"mcpServers": {
"air-choose": {
"command": "node",
"args": ["/absolute/path/to/air-choose/src/server.js"]
}
}
}
Configuration
| Env Variable | Default | Description |
|---|---|---|
AIR_CHOOSE_INTERVAL_MS |
3600000 (1 hour) |
Scheduler re-check interval |
Tools
| Tool | Purpose |
|---|---|
list_providers |
See airline providers, coverage, browser requirements |
resolve_location |
Turn a place name into each provider's own codes |
add_watch |
Add a route to monitor (resolves + runs first check immediately) |
remove_watch |
Stop watching a route |
list_watches |
See everything currently watched |
get_watch_offers |
Current offers for one watch, sorted by price or date |
poll_new_offers |
Return + clear everything new since last call |
force_check |
Re-run a watch's check right now |
Providers
| Provider | Coverage | Auth | Notes |
|---|---|---|---|
| Kupibilet | Worldwide | No | Accepts ISO2 country codes directly |
| Ryanair | Europe | No | Can rate-limit (409) under heavy use |
| SpiceJet | India | Auto token | Calendar view, no exact departure times |
| Jetstar Asia | Asia-Pacific | Browser session | Excluded by default (Akamai) |
Architecture
MCP Client (Hermes / Claude / ...)
| JSON-RPC (stdio)
air-choose server
|-- Tools (8)
|-- Scheduler (hourly)
|-- Provider Adapters (Kupibilet | Ryanair | SpiceJet | Jetstar)
|-- JSON Store (watches | offers | pending)
Example Flow
Agent -> resolve_location({ query: "Moscow" })
Agent -> resolve_location({ query: "Thailand" })
Agent -> add_watch({
originQuery: "Moscow",
destinationQuery: "Thailand",
dateFrom: "2026-07-21",
dateTo: "2026-08-20"
})
-> Server resolves, runs check, returns offers sorted by price
... one hour later, scheduler re-checks ...
Agent -> poll_new_offers()
-> Returns new or cheaper offers found since last poll
Testing
npm test # unit tests
npm run test:e2e # end-to-end (hits live APIs)
npm run test:all # all tests
Legal
Reverse-engineered consumer-facing endpoints, not published partner APIs. Keep the hourly interval reasonable.
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.