Gondola

Travel hacking toolkit for frequent travelers

Gondola is a hotel booking platform built for frequent travelers. Book directly with hotels, earn an average of 16% back in hotel points, credit card points, and 3-7% Gondola Cash back, and let Gondola do the work in the background.

What you can ask

Fetch my travel context and tell me who I am as a frequent traveler. How should I think about my loyalty portfolio?
Example path →

MCP connector

Server URL: https://mcp.gondola.ai/mcp
Login: OAuth
Protocol: MCP 2026-07-28 · io.modelcontextprotocol/tasks
Instructions: Gondola shows hotels with BOTH cash and loyalty-points pricing, plus a cents-per-point (CPP) value, so you can recommend the best-value way to pay. Flights can be award-priced on request: set points: true on when the traveler asks about miles, then call to collect it — award prices come from live airline scrapes, so they arrive on that second call rather than in the search response. Rental cars remain cash-only: never state or imply an award cost for a car.
Hotel questions:
For a named hotel, call with only hotel_id (no dates or occupancy) and present the returned property facts before requesting rates. This is a content-only lookup that does not wait for prices. Reuse details already available in the conversation. Room inventory is not dated availability; do not invent suite or bedroom facts when content is missing. For dated prices and availability, then call ; it may return pending while more sources finish. Content-only questions do not need a rate lookup.
Presenting results to the user:
When you show more than one option side by side (hotels, rooms/rates, loyalty accounts, trips, or a rate comparison), a table with a header row reads better than a list in clients that render Markdown; fall back to aligned plain text where the client does not.
Keep columns tight and lead with the value signal (CPP or deal score). Prose around the table should call out the single best-value pick, not restate every cell.
Using the user's travel context:
returns the user's loyalty programs and elite tiers, home airport, preferred airlines and hotel chains, trip patterns, and stated or learned preferences. It is the best source of who this traveler is, so it is worth reading before making hotel, flight, or car recommendations (for example, default the flight origin to their home airport, and favor programs and chains where they hold elite status).

Tools

Reads data38

Changes data2

Deletes or can't be undone7

Skills

None in the store listings or on the server

What you can ask

Fetch my travel context and tell me who I am as a frequent traveler. How should I think about my loyalty portfolio?

Example path →

MCP connector

Server URL

https://mcp.gondola.ai/mcp

OAuth

Protocol

MCP 2026-07-28 · io.modelcontextprotocol/tasks

Instructions

Gondola shows hotels with BOTH cash and loyalty-points pricing, plus a cents-per-point (CPP) value, so you can recommend the best-value way to pay. Flights can be award-priced on request: set points: true on when the traveler asks about miles, then call to collect it — award prices come from live airline scrapes, so they arrive on that second call rather than in the search response. Rental cars remain cash-only: never state or imply an award cost for a car.

Hotel questions:

For a named hotel, call with only hotel_id (no dates or occupancy) and present the returned property facts before requesting rates. This is a content-only lookup that does not wait for prices. Reuse details already available in the conversation. Room inventory is not dated availability; do not invent suite or bedroom facts when content is missing. For dated prices and availability, then call ; it may return pending while more sources finish. Content-only questions do not need a rate lookup.

Presenting results to the user:

When you show more than one option side by side (hotels, rooms/rates, loyalty accounts, trips, or a rate comparison), a table with a header row reads better than a list in clients that render Markdown; fall back to aligned plain text where the client does not.

Keep columns tight and lead with the value signal (CPP or deal score). Prose around the table should call out the single best-value pick, not restate every cell.

Using the user's travel context:

returns the user's loyalty programs and elite tiers, home airport, preferred airlines and hotel chains, trip patterns, and stated or learned preferences. It is the best source of who this traveler is, so it is worth reading before making hotel, flight, or car recommendations (for example, default the flight origin to their home airport, and favor programs and chains where they hold elite status).

Tools