skimetric

Snow forecasts, lift status, season history, costs and AI ski trip planning for Europe

Community: Submitted by a user or imported; check the owner before granting accessOnlineNo sign-inGlobalFreeRead-only

What it can do

    What data it sees

    Do you need an account

    No: the server works without sign-in

    Snow forecasts, lift status, season history, costs and AI ski trip planning for Europe

    How to connect

    How to connect skimetric to Claude

    The server address to paste into Claude settings:

    https://skimetric.eu/mcp

    1. Open Claude (claude.ai or the desktop app).
    2. Go to Settings → Connectors.
    3. Click Add custom connector.
    4. Paste the server address copied below into Remote MCP server URL and click Add.
    5. In a chat, click + → Connectors and switch the new connector on.

    Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

    Authorization

    After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

    How to connect skimetric to ChatGPT

    The server address to paste into ChatGPT settings:

    https://skimetric.eu/mcp

    1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
    2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
    3. Open Settings → Connectors and click Create.
    4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
    5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
    6. In a new chat open + → Apps/Connectors and enable the connector.

    OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

    Authorization

    On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

    How to connect skimetric to Cursor

    The server address to paste into Cursor settings:

    https://skimetric.eu/mcp

    Fastest: click Open in Cursor below and confirm the prompt.

    Manually:

    1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
    2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
    3. Save the file. The server appears in the MCP list; authorize it there if asked.

    Authorization

    If the service needs sign-in, an authorize button appears next to the server in the MCP list.

    Server tool list (19)

    Raw names from tools/list. Only developers need these.

    find_resortsFind European ski resorts by name and/or ISO-2 country code. Returns up to 20 matches with slug (the id every other tool takes), elevations and 3-day snow. Refine the query rather than paging — there is no bulk catalogue access.
    where_to_skiRank resorts by live conditions, optionally shaped by who is going. Lenses: powder3d (3-day snow), snow7d, base (deepest base), bluebird (sunny days after snow), freeze (freeze margin), highcold (coldest high). Optional country ISO-2, min_summit metres, glacier-only, min_lifts. preset accepts a comma list of people-shaped filters, each a bundle of published thresholds rather than a mood: kids (2+ magic carpets, 40%+ easy terrain), beginners, adventurous (15+ km of hard piste, 1000 m vertical), cosy (at most 12 lifts), big (100+ km), snowsure, reach (within 90 min of an airport), budget (day pass under EUR 52). Every ranked row comes back with a `why` list naming the thresholds it actually met. Top 20 max.
    snow_forecastFull 10-day forecast for one resort (slug from find_resorts), per elevation band (base/mid/summit/glacier): daily snow cm with ensemble p25-p75, temps, freezing level, 48h hourly detail, storm context.
    conditionsLive mountain truth for one resort: lifts/pistes open counts from the operator (dated, fail-closed — absent means no source reported, never a guess), skier thumb reports, and the country's avalanche bulletin.
    season_historyTen winters of measured history for one resort: typical fresh snow, snow days and day/night temps per month (Nov-Apr) and elevation band — the 'is February reliable here' answer. month: 1-12 optional filter.
    mountain_factsWhat a mountain IS: lift count by type, piste kilometres by difficulty, the vertical drop and the nursery lifts — as tagged in OpenStreetMap by its mappers. The 'how big is it', 'how steep', 'anything for a five-year-old' answer. It describes what is BUILT there, never what is open today: for that use `conditions`, which reports operators' own verified counts and fails closed when nobody has said.
    season_fortnightsTen winters split into TWELVE fortnights per elevation band — the granularity a trip is actually booked at, because "February" hides the gap between its halves. Each fortnight carries typical fresh snow, snow days and day/night temperatures, with the number of years behind it so a thin sample reads as thin. band optionally filters to base/mid/summit.
    open_outlookChance the lifts run, per day, for one resort over a date window (default: next 10 days; max 61 days) — a FORECAST about operations, never a report of them. For what is verified open right now use `conditions`. Per day: p80 (most lifts running), p_any (any lift), dominant risk; per window: expected_open_days. Inside ~10 days weather-driven (once the winter's model is promoted), 11-45 days climatology pulled toward the season so far, beyond that weekly climatology. Phrase it to the human as likelihood ("likely", "a 70% chance"), never as fact. start/end: YYYY-MM-DD.
    getting_thereHow to reach a resort. Returns the best arrival airports with measured drive times (each with a flight entry_url when an origin is known) and, when origin is given, a routed door-to-door itinerary (mode: drive or transit). origin: city name, IATA code, or 'lat,lon'. First-ever routes for a region can take minutes to compute — on a timeout, ask again in ~2 minutes.
    trip_costsWhat a ski week costs in one country (ISO-2): median 6-day adult lift pass across its ski areas and the modelled bed price per person/night, with sources and the honest 'modelled vs measured' flags.
    daily_snowToday's Daily Snow editorial: expert-style regional briefs on what the weather is actually doing across the Alps and beyond. Optional country ISO-2 or region filter.
    booking_optionsEverything bookable for a resort stay, no trip required: flights to the best gateway, a stay search shaped to the party and dates, ski rental (local shops when we know them), car, transfer, eSIM. Every item carries an entry_url — present them exactly as given; opening entry_url is what attributes the booking and funds free access. Dates ISO (YYYY-MM-DD); origin: city or IATA.
    open_bookingResolve one entry_url to its live partner destination (the click is recorded at this moment). Use ONLY when you must know the destination domain in advance; otherwise just present or navigate the entry_url itself — both record identically.
    plan_tripStart REAL Skym trip research from a natural-language brief (who, when, vibe, budget — e.g. 'Zermatt or somewhere quieter, 2 adults, mid-January, powder focus'). Research takes ~5 minutes and costs one of 3 daily slots per network. Returns: trip_id + grant (poll with trip_status), and claim_url — GIVE claim_url TO THE HUMAN so the trip lands in their own Skimetric account (opening it as a guest grants view-only; signing in grants editing).
    choose_pathChoose one of a delivered trip's researched options (path id from trip_status options[].id). Reversible: choosing again switches. This activates the LIVING JOURNEY — after it, trip_status returns the requirements checklist (flight, stay, gear, transfer, passes …) with a booking entry_url per step. The choice is announced in the trip's own conversation, so the human who opens claim_url sees it too.
    resort_mediaPictures of a resort for vision-capable agents: approved winter photography, and 'topography' circuit renders — satellite/hillshade art of the ACTUAL runs, so an agent can show its user the terrain. Direct image URLs on skimetric.eu (thumb ~640w for cards, mid ~1280w for heroes, url = original). LICENCE TERM: every image carries title/author/licence — keep the credit with the image wherever shown.
    tell_skymSpeak to Skym inside a planned trip: answer its questions (the requirements' `inputs.question`, e.g. who is travelling, gear sizes), correct an assumption, or ask for a rework ('make it cheaper', 'move it a week'). Skym reads it, updates the plan, and may rework the research — poll trip_status afterwards (allow a minute or two): plan changes land in requirements, Skym's worded reply in skym_says. One message per turn; the trip's human sees the same conversation via claim_url.
    webcamsLive webcams at a resort — what the mountain looks like RIGHT NOW (preview image URLs + links). LICENCE TERM: webcams are provided by windy.com and must be credited as such wherever shown.
    trip_statusPoll a planned trip. While research runs: {status, eta}. Once delivered: the researched options (title, dates, costs), the living journey checklist, and BOOKING entry_urls for every arrangeable step — present entry_urls exactly as given; when booking on the user's behalf navigate through them. Always re-show claim_url to the human.