Find your next source.

Useful for

Use cases llm-inferred

Swiss river baseline screening: nearby BAFU hydro stations and current water levels/temperature for a site (get_water_level, list_hydro_stations) · Swiss weather inputs for a project: MeteoSwiss station list and historical weather for a site period (list_weather_stations, get_weather_history) · Swiss site constraints screening: dam registry lookup by canton for a Swiss site (get_dams_by_canton) · Swiss air quality baseline: NABEL station data and LRV limits near a site (list_air_quality_stations, get_air_quality)

Audience llm-inferred

hydrologists · flood consultants · environmental consultants · GIS analysts · planners

Limitations

Constraints llm-inferred

Covers Switzerland only; no UK data, so not relevant to UK sites · Historical weather data is limited to a window of up to 32 days per request · Data is served from third-party Swiss agencies (BAFU, MeteoSwiss, swisstopo, etc.), so availability depends on those sources · Runs as a community npm/Docker package; maintenance risk sits with a single upstream repository · Loads 79 tools by default, which costs tokens; module filtering is needed for efficiency

Useful for

Use cases llm-inferred

Convert site addresses to coordinates for a desk-based constraints screen · Reverse-geocode survey or monitoring point coordinates to nearest address or locality for report labelling · Quick place-name lookup to confirm the county or district of a study area during scoping

Audience llm-inferred

GIS analysts · environmental consultants · civils engineers · planners

Limitations

Constraints llm-inferred

Designed for small volumes of geocoding; bulk CSV or database work is better served by OpenCage CLI tools or direct API calls · Requires Node.js 20 or higher and a local clone, npm install and build · Desktop config needs an absolute path to the built server file · Setup examples target the claude.ai desktop client; other MCP clients may need different configuration · Rate limits apply to accounts with hard limits; subscription accounts have no hard usage limits · First use requires the user to grant permission in the client

Example tool calls docs-verified

{
  "tool": "geocode-forward",
  "arguments": {
    "query": "Trafalgar Square, London",
    "countrycode": "gb",
    "limit": 1
  }
}

{
  "tool": "geocode-reverse",
  "arguments": {
    "latitude": 51.508,
    "longitude": -0.1281
  }
}

{
  "tool": "get-opencage-info",
  "arguments": {}
}

Useful for

Use cases llm-inferred

Site context mapping: geocode project addresses and map nearby OSM features (watercourses, schools, parking) for a site walkover pack · Travel planning for site visits: route_fetch and get_route_directions between office and client sites · Distance screening: geo_distance and sort_by_distance to rank candidate sites or monitoring locations by proximity to a reference point

Audience llm-inferred

GIS analysts · planners · environmental consultants · civils engineers needing site context

Limitations

Constraints llm-inferred

Rate limits apply: DOCS give defaults of Nominatim 1 rps, Overpass 2 per min, OSRM 100 per min, adjustable by flags · Data is OpenStreetMap-based; completeness and currency of tags are not described in DOCS · No UK-specific datasets, coverage statements or UK data sources are described in DOCS · Local install required: download a binary or build from source (Go 1.24+) · No hydrological, flood, drainage or ground data offered; not a substitute for authoritative UK datasets · Source of price data used by analyze_neighborhood is not described in DOCS · Coverage is inferred from ENTRY regions, not stated in DOCS

Useful for

Use cases llm-inferred

Zonal slope, aspect and elevation statistics over a site boundary from a DEM (geo-terrain dem_zonal) for drainage or civils options appraisal · Reproject and validate site boundary and survey vectors (geo-ops CRS transforms, geometry validation) before exchange with clients · Summer Sentinel-2 scene search over a site bbox (geo-stac) for pre-development land cover context · Species occurrence and OSM feature pulls (geo-biodiversity, geo-vector) as inputs to an ecological desk-study baseline

Audience llm-inferred

GIS analysts · geospatial developers · hydrologists · ecologists

Limitations

Constraints docs-verified

Foundation-model embeddings use a stand-in backend by default; real weights need a configured backend · Credentialed servers (geo-warehouse, geo-commercial-imagery, NOAA CDO, IUCN Red List) need API or cloud credentials · ERA5 is reached via the Open-Meteo archive, not a direct CDS client · Copernicus CLMS covers land-monitoring products, not raw Sentinel · Heavy compute depends on the separate aws-geo-compute peer Power and an AWS account · Without the [pdal] extra, point-cloud output is a local container and not interoperable COPC · Repository is named sample-*, so treat it as a reference implementation and assess maintenance risk

Useful for

Use cases llm-inferred

ERA5 rainfall and temperature history (weather_archive) for FRA or drainage baseline climatology · Forecast rainfall and wind for construction-phase site planning · GloFAS river discharge screening (flood_forecast) at site coordinates before gauge data review · CMIP6 climate projections (climate_projection) for climate-change allowances in drainage strategy

Audience llm-inferred

flood consultants · drainage engineers · hydrologists · planners · GIS analysts

Limitations

Constraints llm-inferred

Docs cover free use for non-commercial purposes only; commercial consultancy use must be checked against Open-Meteo terms · ENTRY auth lists API_KEY, but the docs say no key is needed for Open-Meteo; API_KEY only protects HTTP mode · Model-specific tools are regional (ICON for Germany/Europe, MET Norway for Nordic countries); UK coverage is not named · GloFAS flood discharge is modelled, not gauge observations · Responses over 25,000 characters are truncated; narrow the date range or variables to get full data · Response cache is in-memory only and not persistent · Archive ranges ending within the last 5 days are still being backfilled by Open-Meteo · Tool argument names in example_tool_calls are inferred from Open-Meteo API conventions, not from DOCS · Docs give no maintenance or release-status information

Useful for

Use cases llm-inferred

Reference pattern for grounding AI assistant answers in 3D site or catchment context in a CesiumJS viewer · Prototyping AI-assisted camera control and 3D Tiles/terrain review for planning or drainage schemes · Agent-skill-driven CesiumJS code generation for internal geospatial visualisation tools

Audience llm-inferred

GIS analysts · web developers building CesiumJS apps · BIM and civils engineers reviewing 3D models

Limitations

Constraints docs-verified

Repository is deprecated; active development has moved to cesiumjs-ai-starter-app and cesiumjs-skills · Legacy architecture used a WebSocket bridge needing a separate local server process per user, tied to a single browser tab, and not scalable · No UK-specific or consultancy datasets; coverage is global and it provides integration patterns rather than data

Useful for

Use cases llm-inferred

Flood-zone constraint screening for a project area using fema_nfhl · Soil and farmland siting indicators for drainage or SuDS options using nrcs_soils · Species and critical-habitat screening for early ecological constraints using ipac · Composite constraints map of a project area for early site appraisal using map_composer

Audience llm-inferred

GIS analysts · flood consultants · ecologists · planners · civils and BIM engineers

Limitations

Constraints docs-verified

Screening and research aid only; does not make legal or agency determinations or replace agency or Tribal consultation · Does not guarantee that upstream datasets are complete or current · map_composer queries public GIS services at request time, so results depend on those services being available · census and epa_aqs return no data until their credentials are set

Example tool calls docs-verified

{
  "tool": "ipac",
  "arguments": {
    "roi": "<project-area geometry; parameter name not stated in DOCS>"
  }
}

Useful for

Use cases llm-inferred

Check which layers are loaded in an open ArcGIS Pro map before a desk-study analysis · Count features in a named layer (e.g. Buildings) to sense-check a constraints or asset dataset · Zoom the ArcGIS Pro map to a named layer during a site or model review

Audience llm-inferred

GIS analysts · drainage engineers · civils engineers using ArcGIS Pro · developers extending ArcGIS Pro with MCP tools

Limitations

Constraints llm-inferred

Requires ArcGIS Pro to be running on the same machine with the Add-In loaded, because the named pipe only exists in that session · Only four operations are documented; selectByAttribute, getCurrentExtent and exportLayer are listed as next steps, not implemented · pro.countFeatures returns 0 when the layer name is not found, which is indistinguishable from an empty layer · No retry or timeout logic for IPC is documented (listed as a next step) · Agent Mode needs Visual Studio 2022 17.14 or later · Presented as a sample project; containerised deployment is listed as a future step, not provided

Useful for

Use cases llm-inferred

Flood exposure screening for a French study zone using the risque_inondation recipe · Building density and land-cover baseline for a site in France · Coastal land-pressure analysis from DVF transactions for a French coastal scheme · Print-ready PDF maps and QField data-collection packages for site visits in France

Audience llm-inferred

GIS analysts · flood consultants · planners · hydrologists · civils engineers

Limitations

Constraints docs-verified

Catalogue is France-only (IGN and French national datasets); no UK data sources documented · Production-ready only for single-user local deployments; multi-user mode is beta · Requires Docker and a running QGIS Desktop instance · Default WORKSPACE_AUTH_MODE is permissive: unauthenticated non-loopback calls are served · ign_dem_altitudes is compute-only and not renderable · Jobs over 30 s should use async tools; restart_qgis_engine causes 5-15 s of unavailability · Developed at Cerema Méditerranée; long-term maintenance not documented

Useful for

Use cases llm-inferred

Flood exposure screening for a French site: run risque_inondation to get flood zones and building exposure · Field survey prep: export a QField-ready ZIP with an editable Observations layer for site visits · Land-cover and hydrography baseline for a study area using set_study_zone and smart_load

Audience llm-inferred

flood consultants · GIS analysts · planners · environmental consultants · hydrologists

Limitations

Constraints docs-verified

Production-ready only for single-user local deployments; multi-user mode is beta · Requires a Docker host (docker compose) · Datasets are French national sources (IGN, DVF); no UK-specific data is documented, so UK use is limited · ign_dem_altitudes is compute-only and not renderable · restart_qgis_engine causes 5-15 s of unavailability · Jobs expected to exceed 30 s should use async tools to avoid timeouts · skill://file-exchange (bridge_put/bridge_get) is pending implementation · Default WORKSPACE_AUTH_MODE is permissive: unauthenticated non-loopback calls are served

Useful for

Use cases llm-inferred

Client-site presentation: fly the globe to a project location from a natural-language prompt · Visualising a time-dynamic scene (CZML) for a catchment or asset presentation · Prototyping an LLM-driven 3D GIS viewer before wiring in own UK water or civils data

Audience llm-inferred

GIS analysts · software developers building LLM-driven 3D viewers

Limitations

Constraints llm-inferred

Starter template, not a data source: no datasets are bundled · Chat needs an LLM API key; without one only a plain viewer runs · Terrain and imagery quality depends on the optional Cesium Ion token · Requires Node.js ≥ 22 and npm ≥ 9 for local dev, or Docker · Catalogue kind is 'mcp', but the docs describe an MCP client bridge that consumes external servers; no MCP tools are listed in metadata · Metadata region is Global, with no UK-specific data or coverage described

Useful for

Use cases llm-inferred

Desk-based site context screening: geocode an address, buffer 1-2 km, then pull OSM features (e.g. waterways, buildings) for a constraints check · Preliminary civils or drainage gradient checks: elevation profile along a proposed route or corridor · Terrain screening: sample and zonal stats of a client-supplied GeoTIFF DEM over a site boundary · Drive-time accessibility isochrones for planning or appraisal reports (requires OpenRouteService key)

Audience llm-inferred

GIS analysts · planners · drainage engineers · civils engineers · environmental consultants

Limitations

Constraints docs-verified

Release status: PyPI provides v0.3.0; the v0.4.0 release candidate (newest provider and hardening work) is not yet released · Default public providers have usage policies: Nominatim max 1 request/second, OSRM demo server not for production, Overpass fair use · Isochrones require GEO_MCP_ORS_API_KEY, which is unset by default · Docs are inconsistent on tool count (44 in places, 42 in CLI output) · Roadmap items not yet available: Mapbox/Google geocoding, WebSockets, batch file processing, reproject/clip/mosaic · No UK-specific datasets or hydrological time-series (gauge levels, rainfall) documented · Raster and vector tools operate on local files supplied by the user; no bundled datasets documented

Example tool calls docs-verified

{
  "tool": "geocode",
  "arguments": {
    "query": "Brandenburg Gate, Berlin"
  }
}

{
  "tool": "buffer",
  "arguments": {
    "point": {
      "lat": 52.5163,
      "lon": 13.3777
    },
    "distance_m": 1000
  }
}

Useful for

Use cases llm-inferred

Serve PostGIS drainage or flood-model layers to QGIS and ArcGIS Pro via OGC API Features or GeoServices FeatureServer · Publish survey or site investigation layers as vector tiles for client web maps · Expose PostGIS asset or monitoring data to Power BI or Excel via OData for reporting

Audience llm-inferred

GIS analysts · GIS developers · IT and DevOps teams hosting spatial data

Limitations

Constraints docs-verified

Serves data only; users must load their own layers (PostGIS required, bring your own or use the bundled compose stack) · No versioned v* release tagged yet; nightly container image is the current route · Pro and Enterprise features (e.g. FeatureServer applyEdits, service imports, OIDC SSO) need a signed license · ImageServer, OGC API Coverages and WMTS are Preview · Esri client compatibility is limited to the published parity and certification matrices

Useful for

Use cases llm-inferred

Expose an internal Laravel-hosted data service to AI assistants as MCP tools for a project team · Wrap an existing client REST API that has an OpenAPI spec as MCP tools using make:swagger-mcp-tool · Serve project-specific lookups or calculations from a Laravel app to consultants' AI tooling over Streamable HTTP

Audience llm-inferred

PHP/Laravel developers · backend engineers building internal MCP integrations

Limitations

Constraints llm-inferred

Framework only: the docs describe no geospatial, hydrology or environmental datasets, so the 'gis' category tag is not supported by the docs · Streamable HTTP is the only transport in v2.0.0; SSE was removed from runtime, which conflicts with the SSE mention in the entry description · Requires PHP >= 8.2 and Laravel >= 9.x (Lumen optional), so it needs a full Laravel or Lumen app · v2.0.0 has breaking changes (route-driven setup, setServerInfo); v1.0.0 users must migrate · No authentication built in; the production example depends on app-level middleware · Bundled tools are examples (HelloWorldTool, VersionCheckTool), not data tools

Example tool calls docs-verified

{
  "tool": "greeting-tool",
  "arguments": {
    "name": "Alex"
  }
}

Useful for

Use cases llm-inferred

Canadian hydrometric flow and flood-risk context for a flood risk assessment on a Canadian site · Canadian climate and rainfall normals as inputs to drainage design on Canadian projects · Discovering Canadian open geospatial datasets (CKAN, ArcGIS Hub, WFS) for desk-based site baselines · Canadian exchange-rate or commodity price inputs for project costing

Audience llm-inferred

hydrologists · flood consultants · GIS analysts · environmental consultants working on Canadian projects

Limitations

Constraints llm-inferred

Canada-only data; no UK coverage, so of limited use for terramcp.earth's UK-focused work · Entry description states 266 tools, but DOCS state 295 tools · Depends on many upstream government APIs whose availability and rate limits are outside the server's control · Requires Python 3.12+ and uvx to run · No spatial or temporal resolution or update frequency documented

Useful for

Use cases llm-inferred

Cadastral and OEREB public-law restriction check for a Swiss parcel during site appraisal · Elevation profile along a candidate Swiss route or alignment for civils option appraisal · Contaminated-sites and hazard-map screening for a Swiss canton via geodienste.ch layers · Commune and BFS-number lookup to join Swiss project areas to BFS statistics

Audience llm-inferred

GIS analysts · civils engineers · planners · environmental consultants

Limitations

Constraints llm-inferred

Switzerland-only data; provides no UK coverage, so not applicable to UK projects · ENTRY metadata is inconsistent: it says 13 tools while the docs describe 20; its tool list includes output field names (summary, results, count, match_type, is_error), a parameter name (operation) and names not in the docs (swisstopo_feature_lookup, swisstopo_geodata_download); verify tool names against the running server · Version 0.5.0 per docs badge (pre-1.0): maintenance and API stability risk · Depends on external services (Swisstopo, geodienste.ch, Overpass, OpenPLZ); availability not guaranteed · Zoning result is not legally binding (per docs) · Hosted use in browser requires self-deploying the server (e.g. on Render); no hosted endpoint is documented · Time span and spatial/temporal resolution are not documented

Useful for

Use cases llm-inferred

Swiss project site screening using Zurich district and school-district boundary layers from the geoportal · Drainage and water context for Zurich-area reviews using Lake Zurich level and temperature and UGZ weather station data · Environmental baseline for Zurich assessments using live air quality (NO2, O3, PM10, PM2.5) and climate raster layers · Policy and stakeholder background using council resolutions and parliamentary motions

Audience llm-inferred

environmental consultants · planners · GIS analysts · hydrologists (Zurich-area projects)

Limitations

Constraints docs-verified

Covers City of Zurich data only; no UK coverage · SPARQL tool is not registered by default and its endpoint is not yet productive · Releases before 0.6.0 fail at import on mcp 2.x; use 0.6.0 or newer · HTTP transport has no built-in authentication; set MCP_ALLOWED_HOSTS when binding beyond loopback · Live feeds depend on upstream APIs (e.g. ParkenDD); no rate limits documented

Useful for

Use cases llm-inferred

Cloud-free dry-season NDVI/NDWI baseline for an ecology or BNG habitat assessment · Land-cover training polygons from ESA WorldCover for a site-scale classification · Area-wide vegetation and water-index statistics for a catchment or planning screening report

Audience llm-inferred

GIS analysts · ecologists · hydrologists · planners

Limitations

Constraints llm-inferred

Requires Python 3.10+, a Google Earth Engine account and a Google Cloud project with the Earth Engine API enabled · Earth Engine access is free for research and non-commercial use; commercial consultancy use needs checking against Google's terms, which DOCS do not address · execute_code runs arbitrary Python on the local machine with the user's Earth Engine login; do not expose the server to untrusted clients · Initial earthengine authenticate step opens a browser · Roadmap features not yet available: Dynamic World and Hansen training labels, consensus logic, shapefile/KML/GPX/CSV import, local GeoTIFF backend, OpenStreetMap ingest, STAC backend, change-detection helpers · Land-cover classification workflow needs the optional lulc-engine package · No UK-specific datasets (e.g. river gauges, LiDAR DEM) are documented; these would need to be found in the Earth Engine catalog or added via execute_code · Spatial and temporal resolution and update frequency are not stated · Maintenance status and version history are not documented

Useful for

Use cases llm-inferred

Preliminary hazard screening (earthquakes, volcano alert level, typhoons) for a Philippine overseas project site · River flood outlook for a Philippine city or catchment via get_flood_forecast · Procurement and infrastructure-spending review for a Philippine client bid using PhilGEPS notices · Population and poverty baseline for a Philippine development project area

Audience llm-inferred

hazard and risk consultants · GIS analysts · international development consultants · planners

Limitations

Constraints llm-inferred

Philippines-only data, so not relevant to UK sites; the ENTRY regions field wrongly says United States · Catalogue auth field lists PAGASA_API_TOKEN, which conflicts with the docs' no-API-key statement; verify before use · Many sources are scraped HTML, PXWeb JSON or undocumented APIs, so upstream changes can break tools · Election tools read an archive frozen on 2025-05-16 · query_psa_dataset needs an explicit code for every dimension, refuses 'all' and '*', caps at 1000 cells, and needs describe_psa_dataset first · Upstream outages return upstream_error envelopes, not empty lists; handle that flag explicitly · Heuristic indicators are statistical only; allegations need independent investigation and a second source · Each HDX dataset has its own licence; check license_id before reuse · ENTRY tools field lists status values (success, empty, etc.), not tool names; see docs/tool-reference.md

Useful for

Use cases llm-inferred

Buffer site boundaries and overlay with flood-zone or constraint layers via execute_processing · Reproject and clip client GIS layers to a common CRS for drainage or planning baselines · Batch-export per-feature map sheets and attribute tables to CSV for planning reports

Audience llm-inferred

GIS analysts · flood consultants · drainage engineers · planners

Limitations

Constraints docs-verified

Requires the QGIS 3.22+ desktop application · execute_pyqgis runs inside the QGIS process with AST scan, whitelist and confirmation, not a full sandbox · User skills run in-process with the current user's permissions · LLM features need a configured OpenAI-compatible endpoint and network access · Optional MCP export is off by default, listens on 127.0.0.1 and requires a token · Docs mix current features with historical roadmap notes; feature status follows code wiring · Ships no bundled UK or environmental datasets; works on user-supplied layers · ENTRY tools list includes skills (web_search, gis_data_search, format_results, run via run_skill) and a data file (tool_docs_index.json), not callable tools

Useful for

Use cases llm-inferred

Find and repair broken layer sources across an open ArcGIS Pro project before issuing a report · Summarise attribute counts and statistics per category for a drainage or flood asset layer · Select features by location against a boundary layer and export the result for a planning or EIA report

Audience llm-inferred

GIS analysts · hydrologists · flood consultants · drainage engineers · ecologists

Limitations

Constraints llm-inferred

Requires ArcGIS Pro to be open; the MCP endpoint is served from inside Pro, so the client connection fails when Pro is closed · Requires ArcGIS Pro 3.7 or newer; will not load on older versions · Windows-only host environment (add-in and PowerShell bridge) · Arbitrary arcpy execution (execute_arcpy_code) needs an optional Python bridge inside ArcGIS Pro · Endpoints bound to 127.0.0.1, so not reachable from other machines · Tool parameter schemas are not published in the DOCS, so example arguments must be checked against the live server · Destructive tools (remove_map, delete_features, delete_field, update_features) are marked with a warning in DOCS

Useful for

Use cases llm-inferred

Buffer and clip site boundaries and constraint layers to a study area in QGIS · NDVI and land cover change detection from satellite imagery for habitat or land-use baselines · Hotspot analysis of point records (e.g. incident or survey locations) to target site investigation · Publication-ready study area and classification maps for technical reports

Audience llm-inferred

GIS analysts · ecologists · planners · hydrologists

Limitations

Constraints docs-verified

Code execution is not sandboxed at the Python level; review AI-generated code before running it · Requires a QGIS 3.30+ desktop install (tested on 3.44.2) · Cloud AI providers need internet access and an API key; only Ollama runs fully locally · Licence conflict: ENTRY says 'Other' but the docs state MIT · Provides analysis tools, not datasets or live data feeds; no UK-specific data sources are documented · Google Earth Engine bridge is listed in the architecture but its setup is not documented

Useful for

Use cases llm-inferred

Debugging a Python or TypeScript drainage or hydraulic calculation script with agent-driven breakpoints and debug state inspection · Running GIS or model-processing scripts in a VS Code terminal and returning output to the agent · Refactoring data-processing code for a flood or BNG tool using get_references and rename_symbol to check blast radius · Catching type and lint errors in consultancy tooling with get_diagnostics before handover

Audience llm-inferred

software developers building consultancy tooling · GIS analysts scripting in VS Code · hydrologists and drainage engineers who code their own models

Limitations

Constraints llm-inferred

Local only: bound to 127.0.0.1, not reachable from other machines · No authentication: any other local process can run commands through the endpoint · Terminal execution and output buffering require VS Code shell integration; CMD is not supported natively · Not a data source: no geospatial, hydrological or environmental datasets, so the catalogue category 'gis' does not fit · Catalogue tool list omits terminal_run, which the docs list; pylon.* settings are listed as if they were tools · Version 0.5.0 is installed from a locally built .vsix, so maintenance and release cadence are unclear · Requires VS Code to be open with the extension running

Useful for

Use cases llm-inferred

Print site drawings or report PDFs to an on-site IPP/CUPS printer by copying files into the print-queue mount · Run ad-hoc Python calculations in an isolated sandbox for quick data processing · Schedule recurring agent-driven report prompts with cron and fan results out to chosen channels

Audience llm-inferred

MCP and agent developers · self-hosting home-automation users

Limitations

Constraints llm-inferred

Built as a home-automation agent; no UK water, civils or environmental datasets are described · Requires a multi-service self-hosted stack (qBittorrent, Home Assistant, Redis, Camoufox, and others) · Voice features need Raspberry Pi satellite hardware · Sandbox execution is capped at 60s default and 30min max timeout with 64 KB output · Licence is 'Other'; no open data licence described · Requires external API keys and credentials

Useful for

Use cases llm-inferred

Pre-scope satellite imagery availability (Ready vs Archived) over a site before committing to a download · Find dated scenes over a catchment for flood event or land-cover baseline review · Save recurring study-area searches for reuse in monitoring or EIA baseline work

Audience llm-inferred

GIS analysts · flood consultants · ecologists · planners

Limitations

Constraints llm-inferred

Downloads and cart require a Bhoonidhi login (bhd auth login or env vars) · Archived scenes may need a request on the portal before they can be downloaded · Priced and on-order scenes are only staged to the cart; ordering and payment happen on the Bhoonidhi portal · Background download jobs last only as long as the server process; large downloads should use the standalone bhd CLI · Imagery is under the Bhoonidhi EULA: credit ISRO-IRS and do not resell original products commercially · Archive time span and per-mission resolution are not stated in DOCS · Third-party community project; maintenance commitment not stated

Useful for

Use cases llm-inferred

Pre-visit check of road closures and chain controls for US fieldwork sites in covered states · Haulage and site-access route risk screening for US civils projects using live incidents and closures · Wildfire perimeter and road closure context for US environmental site assessments · Travel-time and toll cost estimates for US consultancy site visits

Audience llm-inferred

environmental consultants (US fieldwork logistics) · civil engineers and site planners (US projects) · GIS analysts · general AI assistant users

Limitations

Constraints llm-inferred

US road data only; no UK coverage · Coverage is uneven across the 37 states, and some states offer only roadwork or nothing usable · California-only detail (CHP dispatch logs, per-lane closures, chain-control levels, CAL FIRE perimeters) · Feed failures are served as last-good data flagged stale, so data can be out of date · Watch-area accounts require approval · Self-hosting is not supported by the author · Premium features are planned for the hosted app · Conditions change faster than feeds; not a substitute for official road information

Useful for

Use cases llm-inferred

Site-visit drive and transit time estimates for mainland China project sites · Geocoding a list of China project addresses to coordinates for GIS layers · POI search for nearby facilities (e.g. schools, hospitals, fuel stations) within a radius of a China site · Reverse geocoding coordinate lists to place names for report labelling

Audience llm-inferred

GIS analysts · planners · site engineers

Limitations

Constraints llm-inferred

China-oriented Amap data; of little use for UK addresses or UK coverage · Transit and POI search need an Amap key and are unavailable without one · Free geocoding (Photon/Nominatim) is unreliable for Chinese addresses and some networks cannot reach it · Transit routes have no true polyline; the map card shows a schematic chain instead · Map cards and the config page render only in the Web client; TUI and headless get text only · Map cards mirror only the current turn's route calls and show at most 3 routes per turn · Docs say this is native DeepSeek Harness tooling, not MCP, so it needs that host; entry metadata labels it kind 'mcp' · Entry metadata lists AMAP_API_KEY as an env var; DOCS store the key in ~/.dsh-map-tools/config.json · Entry metadata lists provider, timeoutMs, maxQps and language as tools; DOCS describe them as config settings · Free routing and geocoding sources are rate-limited and outside the plugin's control · Pre-1.0 (0.x) package; caret version ranges do not roll across minor versions