one-roof
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| EPC_API_KEY | No | API key from the MHCLG Get energy performance of buildings data service. Optional; enables true floor area and £/sqft. | |
| EPC_API_EMAIL | No | Email registered at the MHCLG Get energy performance of buildings data service. Optional; enables true floor area and £/sqft. | |
| UNBLOCKER_API_KEY | No | Scrapfly API key for the commercial unblocker. Required to enable Zoopla and OpenRent adapters in hosted mode. |
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
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| resolve_locationA | Turn free text like 'Clapham', 'near Leeds station', or 'SW11' into structured geography (outcodes, centroid, per-portal identifiers). Call this first; feed its result into search_listings. |
| search_listingsA | Primary tool. Federated live search across every portal (Rightmove, Zoopla, OnTheMarket), deduplicated across portals and enriched with open data (last-sold price, EPC floor area, £/sqft, council tax, flood, crime, schools). Returns a coverage flag so truncation is never silent. |
| get_listingA | Full detail plus full open-data enrichment for one listing, by portal and portal listing id. |
| get_property_intelligenceA | Everything known about an address or UPRN from open data alone — no live listing required. Sold history, EPC (incl. true floor area), council tax band, flood risk, crime density, nearby schools + Ofsted, planning, broadband. Works with zero scraping. |
| market_statsB | Area-level market stats: asking vs achieved prices, £/sqft distribution, rental yields, average time on market, from Land Registry + live listings. |
| compare_listingsB | Normalised side-by-side comparison of several listings on price, £/sqft, EPC, and enrichment. |
| commute_searchA | Filter listings by travel time to a destination — e.g. '<=30 min to Canary Wharf by public transport' — using an isochrone. |
| find_comparablesA | Recent nearby Land Registry sales matched on property type, size, and bedroom count — the comparables a surveyor would use. |
| track_searchA | Persist a search and, on later calls, surface listings that are new or reduced since last check. |
| list_sourcesA | Provenance for the last search: which portal adapters ran, whether each was ok/degraded/blocked, cache age, and whether coverage was complete or truncated. Use this to trust or caveat results. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 10 tools
Each tool has a clearly distinct purpose: location resolution, listing search, single-listing detail, open-data intelligence, market stats, comparison, commute filtering, comparables, persistent search tracking, and data provenance. Even the two listing-related tools (search_listings and get_listing) are separated by cardinality, and get_listing vs get_property_intelligence are differentiated by data source (live portal vs open data only).
Most tool names follow a consistent verb_noun pattern (resolve_location, search_listings, get_listing, compare_listings, commute_search, find_comparables, track_search, list_sources). The only deviation is 'market_stats', which is a noun phrase rather than a verb-led name, making it slightly inconsistent with the rest.
The 10 tools are well within the ideal 3-15 range and each earns its place by covering a distinct aspect of property search, enrichment, comparison, and trust/provenance. There's no bloat or redundancy.
The surface is nearly complete for a property search server: resolving locations, searching, getting details, open-data intelligence, market stats, comparisons, commute filtering, comparables, and provenance. The only gap is that track_search allows persisting a search but provides no way to list or delete tracked searches, which is a minor dead end that agents can work around.