SC-MCP-SERVER
MCP server for AI agents to execute the complete NevioServiceCenter flight booking flow. It provides 10 tools for flight search, cart management, passenger details, seat selection, ancillaries, and booking confirmation with automatic JWT token management.
README
SC-MCP-SERVER — Create Booking Agent
MCP server for AI agents to execute the complete NevioServiceCenter flight booking flow. Provides 10 tools against the live APIM gateway with automatic JWT token management.
Quick Start
cd SC-MCP-SERVER
pip install -r requirements.txt
python check.py # verify all 18 checks pass
python server.py # start in stdio mode
Architecture
┌─────────────────────────────────────────────────────────┐
│ AI Assistant │
│ (Claude / Copilot / Custom Agent) │
└──────────────────────┬──────────────────────────────────┘
│ MCP Protocol (stdio / SSE)
▼
┌─────────────────────────────────────────────────────────┐
│ SC-MCP-SERVER v2.0 │
│ │
│ ┌──────────────────────────────────────────────────┐ │
│ │ 10 Booking Agent Tools │ │
│ │ + 1 Prompt (booking_flow guide) │ │
│ └──────────────────────┬───────────────────────────┘ │
│ │ │
│ ┌──────┴───────┐ │
│ │ APIM Client │ │
│ │ + Token Mgr │ │
│ └──────┬───────┘ │
└─────────────────────────┼────────────────────────────────┘
│ HTTPS + JWT
▼
┌──────────────────────────┐
│ APIM Gateway │
│ (Azure API Management) │
│ │
│ /token │
│ /shop/flights │
│ /create-cart │
│ /checkout/passengers │
│ /services │
│ /seatmap │
│ /seat/services │
│ /cart/retrieve │
│ /checkoutConfirm │
│ /orders/retrieve │
└──────────────────────────┘
Booking Flow (10 Tools)
Token management is automatic — the server fetches, caches (~28 min), and refreshes JWT tokens transparently.
Flow Sequence
SearchFlights → CreateCart → UpdatePassengers → [GetServiceCatalog] → [GetSeatMap]
│ │ │ │ │
SKU IDs checkoutId passenger IDs service SKUs seat grid
│ │ │ │
└──────────────┴───────────────────┴───────────────────┘
│
[AddAncillaries] / [AddSeats]
│
RetrieveCart
│
ConfirmBooking
│
orderId / PNR
│
RetrieveOrder
Steps in [ ] are optional.
Tool Reference
| # | Tool | Endpoint | Purpose |
|---|---|---|---|
| 1 | search_flights |
/shop/flights |
Search flight offers |
| 2 | create_cart |
/create-cart |
Create booking cart from SKU IDs |
| 3 | update_passengers |
/checkout/passengers |
Add passenger identity & contact |
| 4 | get_service_catalog |
/services |
Browse ancillary services |
| 5 | get_seat_map |
/seatmap |
View seat availability |
| 6 | add_seats |
/seat/services |
Assign seats |
| 7 | add_ancillaries |
/seat/services |
Add bags, meals, wifi |
| 8 | retrieve_cart |
/cart/retrieve |
Review cart before confirm |
| 9 | confirm_booking |
/checkoutConfirm |
Create order (PNR) — irreversible |
| 10 | retrieve_order |
/orders/retrieve |
Fetch confirmed order |
Rules
- Thread
checkoutIdfrom step 2 through steps 3–9 - Always call
retrieve_cartbeforeconfirm_booking confirm_bookingis irreversible — confirm with user first- At least one passenger must have
contactDetails(email + mobile) - Dates use
YYYY-MM-DDformat
1. search_flights
| Parameter | Type | Required | Description |
|---|---|---|---|
trip_type |
string | Yes | OW, RT, or MC |
departure_location |
string | Yes | IATA code (e.g. HEL) |
arrival_location |
string | Yes | IATA code (e.g. JFK) |
departure_date |
string | Yes | YYYY-MM-DD |
passengers |
list | Yes | [{"passengerTypeCode": "ADT", "discountCode": "ADT"}] |
fare_type |
string | No | Economy (default), Business, First |
return_date |
string | No | Required for RT trips |
2. create_cart
| Parameter | Type | Required | Description |
|---|---|---|---|
sku_ids |
list[str] | Yes | SKU IDs from search results |
airline_id |
string | No | Default AY |
agent_id |
string | No | Default MCP-Agent |
Key outputs: checkoutId (steps 3–9), passengers[].passengerId (step 3), connections[].flights[].id (step 5)
3. update_passengers
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
passengers |
list[dict] | Yes | Identity + contact details |
4. get_service_catalog
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
promotion_code |
string | No | e.g. SCUISEATP |
5. get_seat_map
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
flight_id |
string | Yes | Flight segment ID from create_cart |
promotion_code |
string | No | e.g. SCUISEATP |
6. add_seats
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
traveller |
list[dict] | Yes | [{id, seat: [{flightId, seat, bundleId?}]}] |
7. add_ancillaries
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
traveller |
list[dict] | Yes | [{id, services: [{id, quantity}]}] |
8. retrieve_cart
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
9. confirm_booking
| Parameter | Type | Required | Description |
|---|---|---|---|
checkout_id |
string | Yes | From create_cart |
10. retrieve_order
| Parameter | Type | Required | Description |
|---|---|---|---|
order_id |
string | Yes | From confirm_booking (e.g. 8LUTXS) |
Configuration
| Variable | Default | Description |
|---|---|---|
APIM_BASE_URL |
https://nevioservicecenterapim.azure-api.net |
APIM gateway URL |
APIM_TOKEN_TTL |
1700 |
JWT token cache TTL in seconds |
MCP_TRANSPORT |
stdio |
Transport: stdio or sse |
MCP_PORT |
3100 |
HTTP port (SSE mode) |
MCP_HOST |
127.0.0.1 |
HTTP host (SSE mode) |
MCP_API_KEY |
(empty) | Bearer token for HTTP auth |
API_REQUEST_TIMEOUT |
30 |
HTTP request timeout in seconds |
Connection Guides
Claude Desktop
{
"mcpServers": {
"sc-booking-agent": {
"command": "python",
"args": ["C:/path/to/SC-MCP-SERVER/server.py"]
}
}
}
GitHub Copilot (VS Code)
Add to .vscode/mcp.json:
{
"servers": {
"sc-booking-agent": {
"command": "python",
"args": ["C:/path/to/SC-MCP-SERVER/server.py"]
}
}
}
HTTP/SSE
python server.py --transport sse --port 3100
{
"mcpServers": {
"sc-booking-agent": {
"url": "http://localhost:3100/sse"
}
}
}
REST API (SSE mode)
# List tools
curl http://localhost:3100/api/tools
# Call a tool
curl -X POST http://localhost:3100/api/tools/search_flights \
-H "Content-Type: application/json" \
-d '{"trip_type":"OW","departure_location":"HEL","arrival_location":"JFK","departure_date":"2026-07-24","passengers":[{"passengerTypeCode":"ADT","discountCode":"ADT"}]}'
Project Structure
SC-MCP-SERVER/
├── server.py # Entry point — 10 tools, 1 prompt, SSE/stdio
├── config.py # APIM + transport configuration
├── check.py # Readiness check (18 checks)
├── requirements.txt # Python dependencies
├── README.md # This file
└── src/
└── tools/
│ └── booking_tools.py # 10 booking tool implementations
└── utils/
├── apim_client.py # APIM HTTP client with auto JWT auth
└── token_manager.py # JWT token fetch, cache, refresh
Token Management
- First API call →
TokenManagercallsPOST {APIM_BASE_URL}/token - JWT extracted from
AuthTokencookie (fallback: response body,Set-Cookieheader) - Cached for
APIM_TOKEN_TTLseconds (default ~28 min) - Auto-refreshes on expiry
- On HTTP 401, invalidates token and retries once with a fresh token
No manual token handling needed.
Dependencies
| Package | Purpose |
|---|---|
mcp>=1.0.0 |
Model Context Protocol framework |
httpx>=0.27.0 |
HTTP client for APIM calls |
uvicorn[standard]>=0.30.0 |
ASGI server (SSE transport) |
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.