MCP HeartbeatTHE MCP OBSERVATORY
Public workspace

Explore

Evidence for your next connection.

Independent measurements of remote MCP servers. Every observation has a timestamp.

Read-only probes

MCP server · Data access / search

Doorfarer

Door-to-door trips: every airport within reach of your home, with the train or bus there.

Checks passed
13 / 22

1 failed · 8 not measured

Median response
411 ms

Median of successful probes over 7 days

30-day uptime
Not measured

Share of countable probes over 30 days, including daily rollups. Auth challenges are excluded.

Discovered tools
5

Tool definitions returned by this server

Independent monitoring

Health & checks

Status history

Outdated check

Last connection measurement is over 24 hours old. Current availability is unknown.

Last checked
Check summary by test group
Test areaPassedFailedUnmeasured
ConnectionCan a client connect and discover tools?
500
Reliability & speedDoes it respond consistently and quickly?
004
Access & authenticationWhat is required to use this server?
102
Tool qualityAre the tool definitions usable by an agent?
400
Publisher identityCan the publisher be traced to its source?
312

Unmeasured means there is not enough evidence to decide. How these checks work

Inspect individual checks 22 recorded

Core

  • Pass
    Reachable

    Latest probe returned HTTP 200 in 411 ms.

  • Pass
    Valid handshake

    protocolVersion 2025-11-25, serverInfo.name doorfarer.

  • Pass
    Lists tools

    tools/list returned 5 tools.

  • Pass
    Current protocol

    Negotiated 2025-11-25. The bar is 2025-06-18 or newer.

  • Pass
    HTTPS

    The endpoint URL is https.

Performance

  • Not measured
    Fast (p50)

    Median latency 411 ms over 1 samples in 7 days. The bar is 1000 ms. Need at least 5 samples.

  • Not measured
    Consistent (p95)

    95th percentile 411 ms over 1 samples in 7 days. The bar is 3000 ms. Need at least 20 samples.

  • Not measured
    Reliable (uptime)

    1 of 1 attempts returned a tool list in 30 days (100%). The bar is 99%. Need at least 30 samples.

  • Not measured
    Stable tool list

    Fewer than two successful probes in 30 days, so tool-list stability is not measured.

Access

  • Pass
    Open access

    tools/list succeeded with no credential, and the record declares none.

  • Not measured
    Standard auth

    Auth is not required, so the OAuth metadata check does not apply.

  • Not measured
    Free tier cited

    No publisher quote has been checked.

Tools

  • Pass
    Tools described

    Every tool description is at least 20 characters.

  • Pass
    Valid input schemas

    Every inputSchema is a JSON Schema object.

  • Pass
    Parameters documented

    41 of 41 parameters have a description (100%). The bar is 80%.

  • Pass
    Safety hints

    Every tool sets readOnlyHint or destructiveHint.

Publisher

  • Pass
    Verified namespace

    Listed in the Official MCP Registry.

  • Pass
    Website matches publisher

    The website host matches the reverse-domain namespace.

  • Fail
    Source repo linked

    No repository URL is on file.

  • Not measured
    Repo owner matches

    This namespace is not an io.github login, so the owner comparison does not apply.

  • Pass
    Recently updated

    Updated 2026-10-30T09:06:46.312731Z, within 180 days.

  • Not measured
    Open-source licence

    No GitHub repository to ask about a licence.

Start using Doorfarer

Connect to this server

Choose your client, copy the configuration, then enable the server in your client’s MCP settings.

Run this command in your terminal, then restart Claude Code.

Claude Code

claude mcp add --transport http doorfarer https://doorfarer.com/mcp

Capabilities

Tools 5

find_group_trips

Finds trips for a group whose people leave from different towns to meet in one place on the same dates: each person's own door-to-door trip (the train, bus or car to their best airport, and the flight), one line per town, with the group's total, the longest journey and the dates that work for everyone. E.g. "Four of us want a week in Lisbon in May: two from New Haven, one each from Ithaca and Burlington." Put each town once in homes with how many people leave from it. For "where should we meet?" pass a region, countries or "anywhere" as destinations with anywhere true. For one home town, use find_trips. A month without a year means the next one coming. With no dates from the group this tool runs no search and returns the question to ask; once they name a time, or say any time suits them (flexible true), it searches. A month is enough to search: don't ask for exact dates. If a search finds nothing, say so and offer a wider window or single-town searches; never quote prices from other sources.

Discovered
find_trips

Finds the best round trips from where someone lives, not just from one airport: it adds the train or bus to every airport they can reach and ranks all options door to door, by price, by total travel time, or balanced. Use it when someone wants flights from their home town, could fly from several airports, or asks for the cheapest or fastest way to get somewhere, e.g. "I live in New Haven: should I fly to Lisbon from JFK, Newark, Boston or Hartford?" Not for a search from one airport already chosen. For a one-way trip set one_way (the trip length is then ignored). Multi-city: search each leg as one_way only when the user gave dates for it; otherwise ask for the dates. Set mode only when the user says how they get to the airport (car, train, bus, any): by default a home in the US compares driving, a taxi to a near airport and public transport, and a home elsewhere uses the train or bus. A named departure airport is an ordinary search: pass it as from_airports, with the home town the user gave. Ask when and for how long first: with no dates from the user this tool runs no search and returns the question to ask (and a form); it searches once the user names a time, or says any time suits them (dates "flexible"). A month or holiday without a year means the next one coming; don't ask which year or to confirm a town. Travellers default to 1 and the trip length to 2-14 nights; the answer's summary names both. Any town or village works. The first search from a town we haven't seen takes about 20 seconds while we look up its train and bus connections; tell the user you're checking the connections from their town to each airport. Later searches from that town take a second or two. Give the home town and where to: airport, city or country codes, a list of them, or a region such as "Iberia" or "Southeast Asia". For "anywhere", a continent ("Asia") or "somewhere warm" it finds the cheapest places itself: pass "anywhere", "abroad" (another country) or the continent, or for warm, sunny or a beach the countries that fit the season with anywhere true; it searches the best 6 cities door to door. `dates` says where the dates came from: "from_user" with depart_from and depart_to, or "flexible" when the user said any time suits them; never pick dates yourself. Example call: {"home": "New Haven", "destinations": ["LIS"], "dates": "from_user", "depart_from": "2027-03-01", "depart_to": "2027-03-31", "nights_min": 5, "nights_max": 9, "travellers": 2}. For a season or a month ("next spring", "in June") pass that window yourself as depart_from and depart_to (spring = 1 Mar to 31 May of the coming year); the answer's searched window says what was searched. The widget already shows every option, so reply in at most 4 sentences with no table: Doorfarer's `verdict` word for word as your first sentence (don't write an assessment of your own), the `headline` (flight + ground = total per person), then end with `close` as written. Never list airports without fares and add no facts about transit, airports or the destination. Don't repeat the table or list airports without fares (with no widget, list the top 3 rows compactly: route, dates, total per person), and give a booking link exactly as returned, as a clickable link (don't shorten or rewrite it). Airports more than 8.5 hours from the door are left out and named in the answer (max_ground_hours changes that: set it only if the user says how far they would travel). Say when a ground leg is an estimate. If the answer's status isn't "ok", tell the user plainly what its note says and don't invent prices; if nothing is found, say so and don't quote prices from other sources. Keep the user's own dates, trip length and places: if nothing is found, say so and offer to widen them, instead of widening on your own. Prices are in the traveller's home currency unless `currency` is given, per person; use the totals the answer gives for a party and don't convert or add up prices yourself; judge a price with the answer's verdict or price_calendar, never with outside benchmarks. Prices come from recent searches on Aviasales; the Book link shows today's price. Booking links are affiliate links: Doorfarer may earn a commission if you book, at no extra cost to you. Say so when you show them, and only then. When the user picks a trip, trip_details gives it step by step with a calendar file; when they ask which month is cheapest or whether a price is good, price_calendar answers: it is the cheapest fare found for each month, or the usual price where none was found (not bookable), while find_trips shows fares seen for the real dates (bookable). If an answer is wrong or confusing, report it with send_feedback.

Discovered
price_calendar

Shows the cheapest month to travel from someone's home town to a place, for each airport within reach: the cheapest fare found for each month (or, where none is found, what that route usually costs) plus the train or bus to the airport and the transfer into the city. Takes a few seconds. Airports further than 8.5 hours from home are left out and named. Use it when the user's dates are flexible or they ask when it's cheapest to go; then call find_trips for the month they choose to get bookable trips. Needs the home town; months as "2027-03", at most 12; for a season or a year pass its months ("next spring" = March to May), and the answer says which months have fares (state only those). Prices per person, in the home currency unless `currency` is given. The widget shows the months: reply in 3-4 sentences, with the best month and what it costs ("cheapest found" or "usual price", as the answer says), and don't quote prices from other sources.

Discovered
send_feedback

Tell Doorfarer's maker where Doorfarer fell short, so it can be fixed. Use it when a find_trips answer was wrong, confusing, or missing something you needed to help, or when the user asks you to pass something on. Write as the assistant that used the tool: what you asked for, what came back, what you expected instead. The report is about Doorfarer only, never about the user: leave out names, contact details, travel plans, reasons for travelling and anything the user said about themselves. Be specific and honest, including when the problem may have been your own call. Pass the search_id from the answer's last line. One report per search. If the user asked you to report something, tell them it's sent.

Discovered
trip_details

Shows one trip from a find_trips answer step by step, door to door: leaving home, the train or bus to the airport, the airport wait, the flight, the transfer into the city and arrival, then the way back, with local times. Also gives the booking link and a calendar file (.ics) the user can add. Pass a trip id from the find_trips answer ("Trip ids, row by row", or a row's `id`); "the best one" is the find_trips answer's top pick. Use it when the user picks a trip or asks how the journey works. Ids are kept for 72 hours. The widget shows every leg, so reply in at most 4 sentences with no table: the summary line, the times that matter (leaving home, the flight, arriving) and the booking link. Don't look up flights, airports or transit elsewhere: our times come from the fare, and a layover is a connection that is part of the trip.

Discovered

Recent measurements

Probe history

View status
Latest stored probes. Raw probe rows are kept for 14 days.
CheckedResultResponseTools
Connected411 ms5