conversational booking — Claude + FareEagle MCP
// User talks naturally. AI handles the booking pipeline.
User: Find me cheap flights from Hyderabad to Goa next Friday
AI: → calls search_flights(from: "HYD", to: "GOI", date: "2026-04-10")
AI: Found 12 flights. Cheapest is IndiGo at ₹3,847 departing 06:15.
Next: Air India at ₹4,290 departing 09:40. Want me to book one?
User: Book the IndiGo one. Name: Rahul Sharma, male.
AI: → calls prepare_booking(trace_id: "abc123", result_index: 0)
AI: Price confirmed: ₹3,847. Locking fare...
AI: → calls create_booking(passengers: [{name: "Rahul Sharma", gender: "M"}])
AI: Booking created. Complete payment here:
https://www.fareeagle.com/book/pay?token=fe_xyz789
Link expires in 15 minutes.
MCP Tools
Nine tools. Three services.
One booking pipeline.
Every tool follows the MCP spec with proper annotations — readOnlyHint, destructiveHint, idempotentHint. Search tools need zero authentication.
Booking Pipeline
Search → Lock → Book → Pay
Three API calls from discovery to payment link. The user completes payment on FareEagle's secure checkout — your agent never handles money.
→
02
Prepare
Lock fare, confirm price
→
03
Create
Submit passengers
→
04
Pay
User pays on FareEagle
→
05
Confirmed
E-ticket issued
Integration
Connect in minutes
MCP server endpoint. No API keys needed for search. Standard JSON-RPC 2.0 over HTTP POST.
// Add to your MCP client config (Claude Desktop, ChatGPT, Cursor, etc.)
// Or in Claude Desktop, click "Add custom connector" and paste the URL.
{
"mcpServers": {
"fareeagle": {
"url": "https://www.fareeagle.com/mcp"
}
}
}
# Streamable HTTP MCP — initialize, then call a tool.
# Most users won't run this directly — Claude Desktop, ChatGPT, etc.
# handle the handshake automatically. Shown for transparency.
# 1. Initialize and capture the session id from the response header
SESSION_ID=$(curl -sD - -X POST https://www.fareeagle.com/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"my-app","version":"1.0"}}}' \
| grep -i mcp-session-id | awk '{print $2}' | tr -d '\r\n')
# 2. Call a tool with that session id
curl -X POST https://www.fareeagle.com/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Mcp-Session-Id: $SESSION_ID" \
-d '{
"jsonrpc": "2.0",
"id": 2,
"method": "tools/call",
"params": {
"name": "search_flights",
"arguments": {
"origin": "HYD",
"destination": "GOI",
"date": "2026-04-10",
"adults": 1,
"cabin_class": "Economy"
}
}
}'
# Using the REST API directly
import requests
response = requests.get(
"https://www.fareeagle.com/api/v1/flights/search",
params={
"from": "HYD",
"to": "GOI",
"date": "2026-04-10",
"adults": 1
}
)
flights = response.json()["results"]
for f in flights:
print(f"{f['airline']} {f['departure']} ₹{f['price']}")
Built for AI
What makes this different
Not a scraper. Not a wrapper. A purpose-built AI booking pipeline with real-time pricing and instant confirmation.
Real-time pricing
Every search hits live inventory. Prices include all taxes and FareEagle markup. No hidden fees, no surprises at checkout.
Price tolerance
Built-in tolerance handles minor price fluctuations between search and booking. ₹200 for flights, ₹500 for hotels. No broken flows.
Secure payment bridge
Your agent never touches payment data. We generate a one-time payment link. Users pay on FareEagle's secure checkout.
MCP spec compliant
Full MCP annotations on all 12 tools. readOnlyHint, destructiveHint, idempotentHint, openWorldHint. Works with Claude, GPT, and any MCP client.
15-minute result window
Search results include expiry timestamps. Agents know exactly when to re-search. No stale pricing, no guessing.
Token replay protection
Every payment token is SHA-256 hashed and checked. No double-charges, no replay attacks.
Get started
Build the next AI travel agent
Zero authentication for search. Add FareEagle's MCP server to your AI agent and start booking travel in India today.
