Skip to main content
Glama
cenogram

cenogram-mcp-server

Official

search_transactions

Read-only

Query 8M+ Polish notarial real estate transactions by location, price, date, area, or type; get address, price, area, and price per m². Use TERYT codes for exact matches.

Instructions

Search Polish real estate transactions from the national RCN registry (8M+ records). Returns transaction details: address, date, price, area, price/m², property type. Call list_locations(search=...) first to resolve a place: prefer the returned TERYT code as teryt= (exact administrative match). Pass a name to location= only when the result flags it as an RCN district (rcn_district) — most TERYT names are not valid location= values and silently return zero rows. Example: search for apartments in Mokotów sold in 2024 above 500,000 PLN. Data notes: marketType is NULL for ~55% of records (notary didn't classify) - filtering by marketType excludes them. ~1.7% of records have no transaction_date. Permalink: every result is shareable on the map. From a result's "id:" line and its "Location: °N, °E" line, build https://cenogram.pl/ceny-transakcyjne?src=mcpstdio#v=1&lat=&lng=&z=16&tx= (drop the °N/°E; lat = the °N number, lng = the °E number) — opens that exact transaction on the map. Omit &tx= for the area only. Field provenance: values are from the notarial deed (RCN) by default; computed values (parcel area summed across plots or converted from hectares, an inferred/reclassified property type) and approximated streets are flagged inline with a neutral [...] note. Location matches TERYT districts only - for neighborhoods (osiedla), use search_by_area instead.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pageNoPage number for pagination (default: 1)
sortNoSort by field (default: date)date
floorNoFloor of the unit (piętro lokalu, residential). Multi-select buckets: exact integers incl. '0' (parter) and negatives e.g. '-1' (basement), 'Nplus' e.g. '10plus' = 10 or more, '0plus' = ground and above, 'unknown' = no floor recorded (NULL). E.g. ['0','1','2'] for ground-to-2nd floor. Building storeys are a different attribute. Without 'unknown', rows with no floor are excluded.
limitNoNumber of results (1-50, default 10)
orderNoSort order (default: desc)desc
roomsNoNumber of rooms (izby) filter, residential units only. Multi-select; '8plus' means 8 or more, 'unknown' = no room count recorded (NULL). E.g. ['2','3'] for 2-3 izby flats. Without 'unknown', rows with no room count are excluded.
terytNoTERYT administrative code(s) for precise area filtering. Comma-separated, max 10. 2-digit (voivodeship), 4-digit (county), 6-digit (municipality), or full precinct code (e.g. '321705_2.0054'). Use list_locations to find codes. More precise than 'location' - avoids name ambiguity.
dateToNoEnd date (YYYY-MM-DD)
streetNoStreet name filter, matched anywhere inside the name (e.g. 'Puławska', 'Aleja Waszyngtona'). Give it in the NOMINATIVE and with its Polish diacritics — matching is literal, so 'Karmelickiej' does not find 'Karmelicka' and 'Marszalkowska' does not find 'Marszałkowska'. Either mistake answers with nothing, which reads exactly like 'no such transactions'.
landUseNoRecorded land-use category of the transaction's land. Multi-select from: gruntyZabudowaneIZurbanizowane (built-up and urbanised), gruntyRolne (agricultural), gruntyLesne (forest), terenyKomunikacyjne (transport), inne (other). 'unknown' = no category recorded for the land (NULL) — a legitimate bucket, never a claim that the land has no use. Values are case-sensitive. E.g. ['gruntyRolne'] for farmland, or ['gruntyZabudowaneIZurbanizowane','gruntyRolne'] to compare developed vs farmland.
maxAreaNoMaximum area in m²
minAreaNoMinimum area in m²
dateFromNoStart date (YYYY-MM-DD)
locationNoLocation name - city (e.g. 'Warszawa', 'Kraków', 'Gdańsk') or district (e.g. 'Mokotów', 'Kraków-Podgórze'). 'Warszawa', 'Kraków', 'Łódź' auto-expand to all sub-districts. Prefer teryt= for exact matches; call list_locations(search=...) to confirm a name is valid here — it flags valid ones as rcn_district.
maxPriceNoMaximum price in PLN
minPriceNoMinimum price in PLN
parcelIdNoExact parcel ID as returned in search results (e.g. '146518_8.0108.27'). Must match exactly - copy from a previous search result's parcel_id field.
floodRiskNoFlood-hazard filter. high = most frequent flooding (~1-in-10-year), medium (~1-in-100-year), low = rarest (~1-in-500-year). Selects ONLY transactions whose land sits in a mapped flood zone; absence of a zone is never asserted as 'safe'. Multi-select; e.g. ['medium','high'] = at least medium risk.
marketTypeNoMarket type: primary (developer) or secondary (resale). ~55% of records have unknown market type and will be excluded when this filter is used.
buildingTypeNoBuilding type filter (PKOB classification). 'unknown' = no type recorded (NULL); without it such rows are excluded (~39% of buildings have no type).
propertyTypeNoProperty type filter
unitFunctionNoUnit/apartment function filter. 'unknown' = no function recorded (NULL); without it such rows are excluded. Garages appear only when 'garage' is selected, not via 'unknown'.
landslideRiskNoLandslide-hazard filter, from official landslide-hazard maps (1:10,000 scale). 'landslide' = the land intersects a mapped landslide area; 'threatened' = an area threatened by mass movements. Selects ONLY transactions whose land intersects a mapped hazard area — an intersection means overlap with a mapped area, not that the parcel itself is a landslide; absence of a zone is never asserted as 'safe'. Multi-select; e.g. ['landslide','threatened'] = any mapped hazard.
ownershipTypeNoOwnership / legal-right type filter (rodzaj prawa do nieruchomości). land_ownership; perpetual_usufruct (użytkowanie wieczyste — covers both registry codes for this right); cooperative_ownership; unit_sale; ownership; unit_ownership_with_appurtenant_right; building_ownership_with_appurtenant_right. 'unknown' = no right recorded (NULL). Multi-select; e.g. ['land_ownership','perpetual_usufruct'] to compare ownership vs perpetual usufruct on undeveloped land.
buildingNumberNoBuilding/house number (e.g. '30', '12A'). Requires location or street to be set.
heritageStatusNoHeritage-listing filter. listed = a protected monument on/at the property's land; zone = the land lies within a protected urban layout or the designated surroundings of a monument. Selects ONLY transactions where a listing was detected; absence of a detection is never asserted as 'not listed'. Multi-select; e.g. ['listed'] = individually listed properties only.
buildingStoreysNoNumber of above-ground storeys of the building. Multi-select buckets: exact non-negative integers (e.g. '1','2'), 'Nplus' e.g. '3plus' = 3 or more, 'unknown' = no storey count recorded (NULL). Recorded ONLY for single-building transactions, so 'unknown' covers BOTH a deed with several buildings (no single storey count exists) and a single building with missing data — never read it as 'a building with no storeys'. This is NOT the floor of a unit (see floor). Without 'unknown', rows with no storey count are excluded.
mpzpDesignationNoMPZP zoning designation filter (exact match, e.g. 'budownictwoMieszkanioweWielorodzinne', 'terenObiektowProdukcyjnychSkladowIMagazynow'). Use 'unknown' for rows with no designation recorded (NULL); distinct from the registry code 'brakMPZPLubWZ' (= 'no plan/WZ' recorded as data).
transactionTypeNoTransaction type filter. For market analysis, ALWAYS specify transactionType to exclude non-market transactions (subsidized, foreclosure, public purpose). ~2% of transactions have unknown type (NULL) and are excluded when this filter is used unless 'unknown' is included.
maxFootprintAreaNoMaximum building footprint (ground-plan) area in m², summed over all buildings of the transaction. Distinct from maxArea, which measures usable floor area (units) or land/parcel area. Set only where every linked building has footprint data, so this bound selects only measured rows — absence means 'not measured', not 'no building'.
minFootprintAreaNoMinimum building footprint (ground-plan) area in m², summed over all buildings of the transaction. Distinct from minArea, which measures usable floor area (units) or land/parcel area. Set only where every linked building has footprint data, so this bound selects only measured rows — absence means 'not measured', not 'no building'.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changedv0.12.0
    • changedInput schema / properties / buildingNumber / description
      Previous value: -"Building/house number (e.g. '251C', '12A'). Requires location or street to be set."New value: +"Building/house number (e.g. '30', '12A'). Requires location or street to be set."
    • addedInput schema / properties / buildingStoreys
      Added value: +{
      +  "description": "Number of above-ground storeys of the building. Multi-select buckets: exact non-negative integers (e.g. '1','2'), 'Nplus' e.g. '3plus' = 3 or more, 'unknown' = no storey count recorded (NULL). Recorded ONLY for single-building transactions, so 'unknown' covers BOTH a deed with several buildings (no single storey count exists) and a single building with missing data — never read it as 'a building with no storeys'. This is NOT the floor of a unit (see floor). Without 'unknown', rows with no storey count are excluded.",
      +  "items": {
      +    "pattern": "^(\\d+|\\d+plus|unknown)$",
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / landUse
      Added value: +{
      +  "description": "Recorded land-use category of the transaction's land. Multi-select from: gruntyZabudowaneIZurbanizowane (built-up and urbanised), gruntyRolne (agricultural), gruntyLesne (forest), terenyKomunikacyjne (transport), inne (other). 'unknown' = no category recorded for the land (NULL) — a legitimate bucket, never a claim that the land has no use. Values are case-sensitive. E.g. ['gruntyRolne'] for farmland, or ['gruntyZabudowaneIZurbanizowane','gruntyRolne'] to compare developed vs farmland.",
      +  "items": {
      +    "enum": [
      +      "gruntyZabudowaneIZurbanizowane",
      +      "gruntyRolne",
      +      "gruntyLesne",
      +      "terenyKomunikacyjne",
      +      "inne",
      +      "unknown"
      +    ],
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • changedInput schema / properties / location / description
      Previous value: -"Location name - city (e.g. 'Warszawa', 'Kraków', 'Gdańsk') or district (e.g. 'Mokotów', 'Kraków-Podgórze'). 'Warszawa', 'Kraków', 'Łódź' auto-expand to all sub-districts. Use list_locations to find valid names."New value: +"Location name - city (e.g. 'Warszawa', 'Kraków', 'Gdańsk') or district (e.g. 'Mokotów', 'Kraków-Podgórze'). 'Warszawa', 'Kraków', 'Łódź' auto-expand to all sub-districts. Prefer teryt= for exact matches; call list_locations(search=...) to confirm a name is valid here — it flags valid ones as rcn_district."
    • addedInput schema / properties / maxFootprintArea
      Added value: +{
      +  "description": "Maximum building footprint (ground-plan) area in m², summed over all buildings of the transaction. Distinct from maxArea, which measures usable floor area (units) or land/parcel area. Set only where every linked building has footprint data, so this bound selects only measured rows — absence means 'not measured', not 'no building'.",
      +  "type": "number"
      +}
    • addedInput schema / properties / minFootprintArea
      Added value: +{
      +  "description": "Minimum building footprint (ground-plan) area in m², summed over all buildings of the transaction. Distinct from minArea, which measures usable floor area (units) or land/parcel area. Set only where every linked building has footprint data, so this bound selects only measured rows — absence means 'not measured', not 'no building'.",
      +  "type": "number"
      +}
    • changedInput schema / properties / street / description
      Previous value: -"Street name filter (partial match, e.g. 'Puławska', 'Trakt Lubelski')"New value: +"Street name filter, matched anywhere inside the name (e.g. 'Puławska', 'Aleja Waszyngtona'). Give it in the NOMINATIVE and with its Polish diacritics — matching is literal, so 'Karmelickiej' does not find 'Karmelicka' and 'Marszalkowska' does not find 'Marszałkowska'. Either mistake answers with nothing, which reads exactly like 'no such transactions'."
  2. First observedv0.5.0

TDQS

A4.4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true and destructiveHint=false, so the safety profile is covered. The description nevertheless adds substantial behavioral context beyond the schema: NULL rates (marketType ~55%, no transaction_date ~1.7%), inline provenance flags, and the fact that bad name matches 'silently return zero rows'. Strong additions, but the core safety signals come from annotations.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with purpose, then usage, then data caveats, permalink, and provenance. For a 31-parameter tool the length is defensible, but the multi-sentence permalink-construction block is dense and leans toward reference-document territory rather than selection guidance.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 31 parameters, no output schema, and no nested objects, the definition supplies the routing rules, the returned fields, the data caveats, and the sibling alternatives an agent needs to call it correctly. Nothing essential for correct invocation is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3 and the schema itself carries most parameter detail. The description adds real value by explaining the teryt-vs-location tradeoff and the silent-zero-row failure mode of incorrect name/diacritic usage, which the schema only partially covers.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb+resource with scope ('Search Polish real estate transactions from the national RCN registry (8M+ records)') and enumerates the returned fields. It explicitly distinguishes itself from siblings by naming search_by_area for neighborhoods/osiedla and list_locations for resolving places, so an agent can route correctly without opening schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives explicit when-to-use routing: 'Call list_locations(search=...) first to resolve a place', 'prefer the returned TERYT code as teryt=', and restricts location= to results flagged rcn_district. It also names the alternative (search_by_area) and the condition that selects it, plus a worked example query.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.