Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
geocode_addressA

Resolve a French address, street, or city to normalized candidates with coordinates, INSEE city code and BAN id (Base Adresse Nationale). Use this first if an address is ambiguous.

property_salesA

List actual notarized property sales (price, date, surface, rooms) recorded by the French tax administration (DVF) around an address, or for a whole commune if only a city is given. Data 2021-2025, no API key.

price_per_m2A

Compute price-per-m² statistics (median, mean, quartiles, per-year evolution) from actual notarized sales around a French address or across a commune. Data 2021-2025.

estimate_propertyA

Transparent comparables-based valuation of a flat or house at a precise French address: weighted median of nearby single-dwelling notarized sales, adjusted to current market level, with every comp and weight returned for audit. Also returns the official rent indicator and gross rental yield.

rent_estimateA

Official modelled asking rents (€/m²/month, charges included) for any French commune: apartments overall, 1-2 rooms, 3+ rooms, and houses. Source: Carte des loyers, Ministère du Logement / ANIL.

property_tax_estimateA

Returns the annual average charge per taxable REI article for taxe foncière on built properties, including published communal, intercommunal, syndicate, GEMAPI and TEOM components. It is an aggregate proxy, not an individual property's tax bill. Source: DGFiP REI via OFGL.

property_reportA

One call, full dossier for a French address: price-per-m² market stats, recent notarized sales, energy diagnostics (DPE), natural & technological risks, commune profile, official rent indicators and an aggregate taxe foncière proxy — plus a comparables-based valuation, gross yield and yield after average property tax when type_local and surface_m2 are provided. Ideal first call when a user asks about a specific property.

dpe_lookupA

Find official energy performance certificates (DPE: energy label A-G, GES label, surface, construction year) filed for a French address. Source: ADEME open data.

natural_risksA

Official risk report for a French address or point: flood, clay shrink-swell, radon, earthquake, industrial sites... Source: Géorisques (Ministère de la Transition écologique).

commune_infoA

Population, postcodes, département, région, surface and center coordinates of a French commune, by name or INSEE code.

reverse_geocodeB

Find the nearest French address for GPS coordinates.

backtest_estimatorA

Replay the comparable-sales estimator on the commune's own notarized sales: every historical sale is valued using only the sales recorded before it (no look-ahead, own deed excluded), then compared with the price actually paid. Returns MAPE, median and 90th-percentile absolute error, signed bias, P25–P75 range coverage and the same metrics by surface band and by year. Use it to say how reliable estimate_property is in this specific market.

cadastral_parcelA

Official cadastral parcel(s) containing an address: unique cadastral id (idu) quoted on deeds, section, number and the official contenance in m². The contenance is the taxed area of the whole parcel, land included — not the dwelling's habitable surface — and the cadastre never records ownership. Source: IGN / DGFiP via API Carto.

urbanism_zoningA

Urbanism zoning and prescriptions applicable to an address, from the Géoportail de l'urbanisme: zoning area label and type (U urban, AU to be urbanised, A agricultural, N natural), the regulation document reference, and the surface / linear / point prescriptions (emplacements réservés, protected areas, alignments…). It states which rules apply, never whether a project is permitted. Source: GPU, DGALN / IGN.

iris_lookupA

Identify the IRIS — INSEE's infra-communal statistical unit, about 2 000 inhabitants — containing an address: 9-character IRIS code, name, type and commune. Join that code to INSEE's IRIS tables for population and Filosofi income. Boundaries from IGN ADMINEXPRESS; this layer carries identity only, no socio-demographic figures.

rent_controlA

Loyer de référence, loyer de référence majoré (the legal ceiling) and minoré, in €/m²/month, for a rent-controlled area — by number of rooms, construction period and furnished status. Covers only the areas whose authority publishes an open grid (Paris, Métropole de Lyon); the response states explicitly when an address is not covered instead of guessing. Sources: Ville de Paris, Métropole de Lyon.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.7/5.0

Scored across 16 tools

Disambiguation3/5

Most tools target distinct official data sources (DPE, risks, cadastre, IRIS), but property_report is an umbrella that overlaps with estimate_property, property_tax_estimate, price_per_m2, dpe_lookup, and natural_risks, and rent_estimate vs rent_control both return rent figures in €/m². Detailed descriptions help, but an agent must read carefully to avoid picking a redundant or wrong tool.

Naming Consistency3/5

All names are readable lowercase snake_case, but there is no consistent verb_noun pattern: geocode_address, estimate_property, and reverse_geocode are actions, while property_sales, natural_risks, commune_info, and cadastral_parcel are noun phrases, and dpe_lookup/iris_lookup put the noun first. The mixed conventions are understandable but not predictable.

Tool Count4/5

16 tools is one over the typical 3-15 well-scoped range, but each tool mostly maps to a distinct official data source (DVF, ADEME, Géorisques, cadastre, urbanism, INSEE) for a broad French real-estate domain. property_report makes some tools redundant, yet the count remains reasonable for the server's comprehensive purpose.

Completeness4/5

The surface covers geocoding, sales, price statistics, comparables-based valuation with backtesting, rents, rent control, tax proxies, energy diagnostics, natural risks, cadastre, urbanism zoning, and commune/IRIS demographics. Minor gaps include aggregate-only tax estimates, no ownership records, and no school or amenity data, but agents can work around these.

Maintenance

ActivityMaintained
ResponsivenessNo issues