suomiatlas

endpoint

https://suomiatlas.com/api/mcp

protocol

http-sse ·2025-06-18

authentication

none observed

public key

none — nobody has proven they own this listing · is it yours? claim it

karma

0 · newcomer

Is suomiatlas live?

Yes — it answered the hub's last check (checked 3h ago). It answered 100% of checks over the last 30 days.

Is suomiatlas free to use?

Yes — the hub reached it with no key and no payment.

What tools does suomiatlas have?

10 tools: get_air_quality, get_nearby_places, compare_areas, rank_areas, find_similar_areas, get_area_report, get_place_history, search_places, ….

Is suomiatlas safe to connect?

The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

• list_variables reads open 13h ago

  The catalogue of every metric Suomiatlas publishes — population, income, education, employment, buildings and dwellings, housing sale prices, crime rates, composite scores — grouped by theme, with codes, units and types. Call this to find the right `variableCode` for rank_areas or get_place_stats. Pass `variableCode` to get the years that metric covers instead; coverage differs per metric (Paavo 2010–2024, housing prices 2009–2025, air quality per municipality from 2019 with fewer stations in the early years). Air-quality codes (aq_pm25_avg, aq_pm10_avg, aq_no2_avg, aq_o3_avg) are municipality-grain and not in this catalogue; pass them to rank_areas directly. An empty year list for a code means it is unpublished, not that the metric is zero.

  mcp-tool

  123456789101112131415

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "properties": {
      "group": {
        "type": "string",
        "description": "Filter to one theme, matched case-insensitively, e.g. \"income\", \"crime\"."
      },
      "variableCode": {
        "type": "string",
        "description": "Return the years available for this metric rather than the catalogue."
      }
    },
    "additionalProperties": false
  }

  arguments 15 lines copy
• get_air_quality reads unknown 13h ago

  Measured annual-mean air quality (PM2.5, PM10, NO₂, O₃) for a Finnish postal area or municipality, from Finnish Meteorological Institute and municipal monitoring stations, each compared against the WHO 2021 guideline and the EU limit. Answers "is the air clean in Vantaa?" with a verdict rather than a bare number, and returns each pollutant's multi-year `series` beside its latest reading. Coverage is sparse: Finland has ~160 stations, so a pollutant with no nearby station is reported as a gap — report that as unmeasured, never as clean.

  mcp-tool

  1234567891011121314151617181920212223

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "kind": {
        "enum": [
          "postal_area",
          "municipality"
        ],
        "type": "string",
        "description": "Municipality gives the median over its stations; postal area uses the nearest station."
      },
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Place name or code, e.g. \"Vantaa\", \"00560\""
      }
    },
    "additionalProperties": false
  }

  arguments 23 lines copy
• get_nearby_places reads unknown 6h ago

  What is physically inside one Finnish postal area: public transport stops (bus, tram, metro, train, ferry) from Digitransit; daycare centres from the Suomi.fi service catalogue (PTV), with OpenStreetMap adding only those with no PTV daycare within 250 m; and schools, higher and further education, health clinics, pharmacies, supermarkets and libraries from OpenStreetMap; and swimming pools and beaches from LIPAS, the national register of sports facilities. Answers "what is there in Tikkurila?" or "is there a school in 99980?" with counts and names. With `withinKm` (up to 5) it also looks beyond the area: places within that straight-line distance of the area's boundary, nearest first per category with their distance — one call for "is there a pool or a beach near 00250?". Postal areas only; a city name returns its areas to choose from. Transit counts are platforms (most stops have one per direction); `names` groups platforms by stop name. A zero for an OpenStreetMap category means none is mapped there, not that none exists; a null count means that source has not been loaded; a null name is a place its source has no name for. This says nothing about travel times or service frequency. Licensed per source: the OpenStreetMap places and the daycare layer are ODbL 1.0, the pools and beaches CC BY 4.0; credit "© OpenStreetMap contributors", Digitransit, the Suomi.fi service catalogue and "Jyväskylän yliopisto, Lipas Liikuntapaikat.fi" when reproducing it.

  mcp-tool

  123456789101112131415161718192021222324252627282930313233343536373839404142434445

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Postal area name or 5-digit code, e.g. \"Tikkurila\", \"00530\""
      },
      "withinKm": {
        "type": "number",
        "maximum": 5,
        "minimum": 0.1,
        "description": "Also look this many km beyond the area's boundary (straight-line, 0.1 to 5), and answer with the nearest places and their distance."
      },
      "categories": {
        "type": "array",
        "items": {
          "enum": [
            "bus",
            "tram",
            "metro",
            "train",
            "ferry",
            "kindergarten",
            "school",
            "higher_education",
            "health_clinic",
            "pharmacy",
            "supermarket",
            "library",
            "swimming_pool",
            "beach"
          ],
          "type": "string"
        },
        "minItems": 1,
        "description": "Only these categories; defaults to all fourteen."
      }
    },
    "additionalProperties": false
  }

  arguments 45 lines copy
• compare_areas reads unknown 20h ago

  Side-by-side statistics for up to 12 Finnish postal areas, with national benchmarks and percentiles. Accepts names or codes. A null value or percentile is suppressed or unpublished for that area and year (small-area confidentiality), not zero.

  mcp-tool

  123456789101112131415161718192021222324

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "places"
    ],
    "properties": {
      "year": {
        "type": "integer",
        "description": "Statistics year; defaults to the newest."
      },
      "places": {
        "type": "array",
        "items": {
          "type": "string",
          "minLength": 1
        },
        "maxItems": 12,
        "minItems": 2,
        "description": "Postal area names or 5-digit codes, 2 to 12 of them, e.g. [\"Hervanta\", \"Tikkurila\", \"02100\"]."
      }
    },
    "additionalProperties": false
  }

  arguments 24 lines copy
• rank_areas reads unknown 20h ago

  Rank Finnish postal areas by any metric — highest or lowest income, population, unemployment, education, housing prices — nationally or within one municipality or region. Crime rates (crime_total_rate and friends) and air quality (aq_pm25_avg, aq_pm10_avg, aq_no2_avg, aq_o3_avg — annual means in µg/m³) rank municipalities instead, since both are published only at that grain; the aq_ codes are not in list_variables, pass them directly. For air quality order="bottom" gives the cleanest municipalities first and order="top" the most polluted, so "which municipality has the cleanest air?" is rank_areas with aq_pm25_avg and order="bottom" — get_air_quality answers for one place only. A municipality absent from an air-quality or crime ranking has no published value there: report it as unmeasured, never as clean or safe. A null in `additionalValues` means that metric is unpublished for that area, not zero. For gender_balance_20_39 the useful question is usually which areas are closest to an even split, not which are most skewed: pass order="balance" for that, since both a high and a low ratio mean imbalance. Get valid metric codes from list_variables. `near` plus `withinKm` narrows the ranking to areas within a straight-line radius of another place. That is distance over the ground, never a travel or commute time. To screen areas on several metrics at once — "above-average family share and below-average housing prices" — rank by one metric and pass the rest as `includeVariables`; that is a single call. Do not call get_place_stats per area to collect a second metric.

  mcp-tool

  12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "variableCode"
    ],
    "properties": {
      "near": {
        "type": "string",
        "description": "Keep only areas within `withinKm` of this place — a postal area or municipality, by name or code, e.g. \"Tampere\", \"00100\". Distance is centre to centre in a straight line."
      },
      "year": {
        "type": "integer",
        "description": "Year; defaults to the newest with data."
      },
      "limit": {
        "type": "integer",
        "maximum": 100,
        "minimum": 1,
        "description": "Rows to return, 1–100 (default 10)."
      },
      "order": {
        "enum": [
          "top",
          "bottom",
          "balance"
        ],
        "type": "string",
        "description": "Highest first (default), lowest first, or — for a metric whose ideal value is a specific number rather than an extreme — closest to that value first. \"balance\" is currently valid only for gender_balance_20_39, whose target is an even 1.0; asking for it on any other metric is an error. For aq_* metrics \"bottom\" is cleanest first."
      },
      "scope": {
        "type": "string",
        "description": "Limit to one municipality or region by name or code, e.g. \"Espoo\", \"Uusimaa\". Not applicable to crime or air-quality rankings, which are national lists of municipalities — use `near` + `withinKm` to narrow those."
      },
      "withinKm": {
        "type": "number",
        "description": "Radius in kilometres around `near`. Required with `near`, and only with it.",
        "exclusiveMinimum": 0
      },
      "variableCode": {
        "type": "string",
        "minLength": 1,
        "description": "Metric code from list_variables, e.g. \"hr_mtu\", \"unemployment_rate\", or an air-quality code such as \"aq_pm25_avg\"."
      },
      "includeVariables": {
        "type": "array",
        "items": {
          "type": "string"
        },
        "maxItems": 5,
        "description": "Extra metric codes to report for each ranked area, as `additionalValues` on every row. Each metric's own year is given in `includedVariables`, since the sources run to different years. Not available for municipality-grain rankings (crime, air quality)."
      }
    },
    "additionalProperties": false
  }

  arguments 55 lines copy
• find_similar_areas reads unknown 20h ago

  Postal areas most like a given one across its whole statistical profile — income, age structure, education, housing and employment together, not one metric. Useful for "where else in Finland feels like this neighbourhood?". Similarity is statistical, not geographic: the results are usually spread across the country. An area whose statistics are suppressed shares too few dimensions to be compared and is left out, so absence from the list is not dissimilarity — it is unmeasured.

  mcp-tool

  12345678910111213141516171819202122232425

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "year": {
        "type": "integer",
        "description": "Year; defaults to the newest."
      },
      "limit": {
        "type": "integer",
        "maximum": 50,
        "minimum": 1,
        "description": "Neighbours to return (default 10)."
      },
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Postal area name or 5-digit code, e.g. \"Hervanta\", \"00530\"."
      }
    },
    "additionalProperties": false
  }

  arguments 25 lines copy
• get_area_report reads unknown 6h ago

  The download link for one postal area's printable PDF report (composite scores, trends and the full statistics table). Returns a URL and never the file itself — the report renders the same data the other tools return, so offer the link when someone wants the document, and answer questions from get_place_stats instead. Downloading requires an Area Pass for that area; without one, the link still comes back and `locked` says what it costs. Reports exist per postal area only, not per municipality.

  mcp-tool

  12345678910111213141516171819202122232425262728

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "year": {
        "type": "integer",
        "description": "Statistics year; defaults to the newest."
      },
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Postal area name or 5-digit code, e.g. \"Hervanta\", \"00560\""
      },
      "language": {
        "enum": [
          "fi",
          "sv",
          "en"
        ],
        "type": "string",
        "description": "Report language: fi (default), sv or en."
      }
    },
    "additionalProperties": false
  }

  arguments 28 lines copy
• get_place_history reads unknown 25h ago

  Time series for one Finnish postal area or municipality, with the direction and size of change computed for each metric so the trend does not have to be eyeballed. Answers questions like "have apartment prices been rising in Espoo?" and "is the population growing?". Defaults to housing sale prices and population. For a postal area it also returns `trendMetrics`: precomputed 1/5/10-year indicators — population growth, income momentum, aging velocity, family share change, dwelling count and stock change, education-level change, and unemployment relative to the national change. They describe how the area has changed ("has the population grown? have incomes risen faster than nationally?"), not why or what comes next, so do not present them as a verdict such as "gentrifying" or "declining". Each carries `history`, its per-year values for up to 10 years, when the backend has computed them. Series points are index-aligned to `years`: a null point is a year the source did not publish for this area, and `trend` is null when fewer than two points have values or the series starts at zero — call that too sparse to read, not flat.

  mcp-tool

  123456789101112131415161718192021222324252627282930

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "kind": {
        "enum": [
          "postal_area",
          "municipality"
        ],
        "type": "string",
        "description": "Force a grain."
      },
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Place name or code, e.g. \"Espoo\", \"Hervanta\", \"00560\""
      },
      "variables": {
        "type": "array",
        "items": {
          "type": "string"
        },
        "description": "Variable codes to chart (see list_variables). Defaults to housing prices and population."
      }
    },
    "additionalProperties": false
  }

  arguments 30 lines copy
• search_places reads unknown 6h ago

  Resolve a Finnish place name to the code every other Suomiatlas tool takes. Handles postal areas (e.g. Toukola, Töölö), municipalities (Espoo, Akaa) and regions (Uusimaa). Diacritics are optional: "toolo" finds "Töölö". Call this first whenever the user names a place rather than a code. Many Finnish neighbourhood names are shared across cities, so several results is normal — present the choices instead of assuming the first one. Postal-area results carry an inhabitant count with the year it was published for; if you quote it, quote the year with it. `population` is null for municipalities and regions — the count is carried per postal area only — not a sign the place is empty.

  mcp-tool

  123456789101112131415161718192021222324252627282930

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "query"
    ],
    "properties": {
      "kind": {
        "enum": [
          "postal_area",
          "municipality",
          "region"
        ],
        "type": "string",
        "description": "Restrict to one grain. Omit to search all three."
      },
      "limit": {
        "type": "integer",
        "maximum": 50,
        "minimum": 1,
        "description": "Maximum results (default 10)."
      },
      "query": {
        "type": "string",
        "minLength": 1,
        "description": "Place name or code, e.g. \"Töölö\", \"Espoo\", \"00560\""
      }
    },
    "additionalProperties": false
  }

  arguments 30 lines copy
• get_place_stats reads unknown 13h ago

  Current statistics for one Finnish postal area or municipality: population, age, income, employment, education, housing sale prices (€/m²) and municipal crime rates, with national/regional percentiles where available. Accepts a name or a code and resolves it automatically; if the name is ambiguous it returns the candidates instead of guessing. Answers questions like "how many people live in Hervanta?", "how many live in Akaa?" and "how much is an apartment in Matinkylä?". Returns a headline set of metrics by default — pass `variables` to fetch specific codes from list_variables. A postal area's result also includes `ratios[]`: household composition and jobs-to-residents ratios compared against Finland as a whole (see each entry's `nationalBasis` and the response's `notes`). A null value is a figure Statistics Finland suppressed (small-area confidentiality) or never published for that vintage, and a null percentile means one could not be computed — report both as unpublished, never as zero; `ratios[].areaValue` is null when the area's denominator is under 30 or an input is unpublished.

  mcp-tool

  123456789101112131415161718192021222324252627282930313233343536373839404142

  {
    "type": "object",
    "$schema": "http://json-schema.org/draft-07/schema#",
    "required": [
      "place"
    ],
    "properties": {
      "kind": {
        "enum": [
          "postal_area",
          "municipality"
        ],
        "type": "string",
        "description": "Force a grain when a name exists at both, e.g. Helsinki the municipality."
      },
      "year": {
        "type": "integer",
        "description": "Statistics year; defaults to the newest available."
      },
      "place": {
        "type": "string",
        "minLength": 1,
        "description": "Place name or code, e.g. \"Hervanta\", \"Akaa\", \"00560\""
      },
      "variables": {
        "type": "array",
        "items": {
          "type": "string"
        },
        "description": "Extra variable codes to include beyond the headline set (see list_variables)."
      },
      "includeAllVariables": {
        "type": "boolean",
        "description": "Return all ~131 variables. Large; only use when the headline set is insufficient."
      },
      "includeCompositeScores": {
        "type": "boolean",
        "description": "Include composite 0-100 scores with their component percentiles (postal areas only)."
      }
    },
    "additionalProperties": false
  }

  arguments 42 lines copy

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.