Data offered docs-verified: live aircraft · ships · satellites · earthquakes · traffic · public cameras
Coverage llm-inferred: Global (globe-wide); public camera feeds limited to Austin, London, California, Finland and Delaware
Update frequency docs-verified: Live or regularly refreshed; per-feed intervals not documented
Access docs-verified: Free and open source (licence: Other, per metadata); runs locally in a browser. Optional keys: Cesium ion token (free, eligible personal non-commercial use), Google Maps key (paid, metered), OpenAI key (voice control only)

Useful for

Use cases llm-inferred: Desk-based site context overview (terrain, imagery, nearby live signals) before a site visit · Weather context review for an event window using forecast wind and observed radar replay · Visual briefing material for client and stakeholder presentations using scene capture
Audience llm-inferred: GIS analysts · flood consultants (visualisation only) · planners · civils/BIM engineers (visualisation only)

Limitations

Constraints llm-inferred: Traffic is simulated along real roads from aggregate location data, not measured · CCTV camera poses and rocket launch trajectories are coarse estimates · Public camera feeds only over listed regions · Google Maps and Cesium ion features are subject to provider terms and quotas · Voice control requires an OpenAI key · Bundled Nepal flood data is non-commercial only · Licence listed as Other, so reuse terms need checking before commercial use · Older versions no longer work with OpenStreetMap data sources and need updating

gis civils-bim

Globalhealthy

mcp

83/100Currency A

tidewave_rails ↗

MCP server with runtime-level tools for Rails development

What's in it

Data offered docs-verified: runtime code evaluation in the running Rails app (project_eval) · live app logs · development database query results (execute_sql_query) · source file and line locations of classes, modules and methods, including gem dependencies (get_source_location) · documentation for classes, methods and constants pinned to Gemfile.lock gem versions (get_docs)
Formats docs-verified: MCP streamable HTTP
Access llm-inferred: Open source under the Apache License 2.0; no credentials documented in README. Requires a local Rails development app with the tidewave gem added to the development group.

Useful for

Use cases llm-inferred: Reproduce a failing code path against live app state with project_eval when debugging a Rails web tool · Confirm that a create action saved expected records by querying the development database with execute_sql_query · Check method signatures in the exact gem versions locked in Gemfile.lock with get_docs before generating code · Jump to the defining file and line of a gem method with get_source_location when extending a Rails app
Audience llm-inferred: Ruby on Rails developers · web application developers building internal tools for consultancies

Limitations

Constraints llm-inferred: Development use only; not meant for production · Raises an error in any environment where code reloading is disabled · Requires a running Rails app with the tidewave gem in the development group · Accepts requests from localhost only by default · Toolbar may not appear if compression middleware runs after Tidewave · Multiple hosts or subdomains need *.localhost and session configuration changes · Provides no routes or associations tools by design · Contains no UK water, civils or environmental datasets; it is a developer tool, not a data source

environment hydrology

Globalhealthy

mcp

83/100Currency A

139-tool MCP server for real-time global intelligence from free public APIs: markets, SEC filings, conflict, military, aircraft-cohort early warning, refugee and migration flows, cyber, climate, news, and 30+ domains. Cited AI situation briefs, geofences with change detection, live SSE dashboard. MIT, no paid API keys.

What's in it

Data offered llm-inferred: earthquakes (USGS) · wildfire hotspots (NASA FIRMS) · disaster alerts (GDACS) and environmental events (NASA EONET) · conflict events (ACLED, UCDP) · refugee and IDP displacement (UNHCR) · FX rates (ECB) · macro and economic indicators (FRED, World Bank, EIA) · internet outages and cloud service status (Cloudflare Radar, cloud status) · severe weather alerts (NWS, Meteoalarm) · tropical cyclones (NHC, JTWC) · volcano activity (Smithsonian GVP) · news feeds and trending keywords (RSS, GDELT)
Coverage docs-verified: Global
Update frequency docs-verified: Default 5-minute interval when the collector runs as a daemon; per-tool refresh rates not stated
Formats llm-inferred: MCP stdio · PDF · HTML · markdown · SSE (live dashboard)
Access docs-verified: Free public APIs with no paid subscription; MIT licence. Several tools need free API credentials set as env vars (ACLED_ACCESS_TOKEN, CLOUDFLARE_API_TOKEN, EIA_API_KEY, FRED_API_KEY, NASA_FIRMS_API_KEY, OPENSKY_CLIENT_ID, OPENSKY_CLIENT_SECRET)

Useful for

Use cases llm-inferred: Hazard screening for a client site: recent earthquakes, wildfire hotspots and GDACS disaster alerts for context in an environmental or civils risk note · Severe weather context for a project briefing using NWS and Meteoalarm alerts where the site falls in US or EU coverage · Macro and cost-escalation inputs for civils and infrastructure budgets using FRED, World Bank and EIA energy price series
Audience llm-inferred: environmental consultants · civils and infrastructure risk analysts · planners · GIS analysts · AI agent developers

Limitations

Constraints llm-inferred: Global geopolitical, market and security focus; no UK hydrology, flood, rainfall or ground-model datasets are listed in DOCS · Severe weather coverage is limited to US (NWS) and EU (Meteoalarm); UK coverage is not listed in DOCS · Requires local install of a Python package (pip install -e .) and Python 3.11+; not a hosted service · PDF reports need native dependencies, e.g. pango on macOS · A source that fails three times in a row is paused for 5 minutes by its circuit breaker · Several tools need API credentials; per-source rate limits and free-tier quotas are not stated in DOCS · Source maintenance risk: the server aggregates 30+ third-party public APIs that can change or disappear

climate gis

Globalhealthy

skill

82/100Currency A

geosql ↗

A Claude/Codex skill for geospatial data

What's in it

Data offered docs-verified: Spatial SQL generation and execution against PostGIS, BigQuery, Snowflake and Wherobots · Overture Maps shares on BigQuery and Snowflake · Warehouse table/column discovery · Map rendering of query results via optional Dekart CLI
Coverage llm-inferred: Global (per ENTRY regions; actual data coverage depends on the connected warehouse)
Access docs-verified: Free and open, MIT licence (per ENTRY). No SaaS account needed; runs locally or self-hosted. Warehouse access uses local CLI authentication (bq, snow, dekart); the optional Dekart CLI is installed on request.

Useful for

Use cases llm-inferred: Screening private warehouse layers for proximity to road or utility networks (e.g. features within 5 km of a motorway) for constraint mapping · Sanity-checking polygon area and line length of GIS query outputs before using them in a report · Joining site or asset points to Overture Maps building or road layers for baseline context · Rendering spatial query results as a map for client or peer review
Audience llm-inferred: data scientists · analysts · GIS analysts · spatial data engineers

Limitations

Constraints llm-inferred: Requires access to a supported warehouse (PostGIS, BigQuery, Snowflake or Wherobots) and its local CLI credentials · BigQuery queries are capped at a 10 GiB scan by default · Benchmark suite covers only London, Berlin and Paris cases (8 assertions) · Map-feedback loop needs the optional Dekart CLI · ENTRY notes no credentials documented in README, while the docs rely on local bq/snow/dekart auth · Not UK-specific; no UK datasets are named in the docs · Version check telemetry is on by default (disable with DO_NOT_TRACK=1)

gis

Globalhealthy

mcp

81/100Currency A

mcp-server ↗

Mapbox Model Context Protocol (MCP) server

What's in it

Data offered docs-verified: geocoding and reverse geocoding (addresses and place names to coordinates and back) · POI search · multi-modal routing (driving, walking, cycling) with real-time traffic · travel time matrices · route optimisation for multiple stops · map matching of GPS traces · isochrones (time or distance reachability) · interactive map rendering (render_map_tool) · static map images · offline distance, area, bearing, buffer, centroid and points-within-polygon calculations
Coverage docs-verified: Global (worldwide)
Formats llm-inferred: MCP (Node.js server; hosted endpoint https://mcp.mapbox.com/mcp) · PNG · JPEG · GeoJSON (input to render_map_tool)
Access docs-verified: API key: Mapbox access token required (env MAPBOX_ACCESS_TOKEN); free Mapbox account signup; server code MIT licence

Useful for

Use cases llm-inferred: Travel-time isochrones and drive or walk reachability from a depot or site for survey and sampling crew scheduling · Travel-time matrix and multi-stop route optimisation for multi-site monitoring rounds such as river or outfall sampling · Static map images of site location, catchment context and sampling points for report figures · Offline buffer, area and point-in-polygon checks on site boundaries or consultation zones
Audience llm-inferred: GIS analysts · civil engineers (site logistics) · environmental consultants · planners

Limitations

Constraints llm-inferred: Requires a Mapbox access token (MAPBOX_ACCESS_TOKEN) · Global, general-purpose location platform; no UK hydrology, flood, rainfall or environmental datasets among the listed tools · Geocoding, POI, routing and isochrone tools call Mapbox APIs; usage limits and pricing are not stated in DOCS · Interactive render_map_tool works only in MCP Apps clients (e.g. Claude Desktop, Claude Code, VS Code with GitHub Copilot, Goose); use static_map_image_tool elsewhere · Resource fallback tools (resource_reader_tool, category_list_tool) are off by default; enable CLIENT_NEEDS_RESOURCE_FALLBACK=true for clients without resource support · Tool argument schemas are not included in DOCS

gis

Globalhealthy

mcp

80/100Currency A

geo-mcp-servers ↗

A curated, tracked list of Model Context Protocol (MCP) servers for geospatial, GIS, mapping, and earth-observation work — geocoding, routing, PostGIS, STAC/imagery, QGIS/ArcGIS, weather, and the commercial location platforms.

What's in it

Data offered docs-verified: catalogue of geospatial MCP servers covering geocoding, routing, PostGIS, STAC/imagery, QGIS/ArcGIS, weather, and commercial location platforms
Formats llm-inferred: Markdown (README tables) · YAML (servers.yaml source)
Access docs-verified: Catalogue licensed MIT (ENTRY metadata). Listed servers range from open to paywalled: open, free, commercial, or paywalled, per the docs legend.

Useful for

Use cases llm-inferred: Find a geocoding or reverse-geocoding MCP server to convert site addresses to coordinates for a planning appraisal · Identify a PostGIS or STAC/imagery MCP server to query spatial datasets for catchment or land-use analysis · Locate a weather MCP server to pull rainfall or climate context for a drainage strategy desk study
Audience llm-inferred: GIS analysts · hydrologists · drainage engineers · flood consultants · planners · developers building MCP-based geospatial tooling

Limitations

Constraints llm-inferred: Lists servers rather than providing data itself · Server health varies: 18 stale and 2 pending at the time of the health check · Health status reflects only repository push activity within 12 months · Some entries have URLs still to be confirmed · Access tiers differ by server, and many are commercial or paywalled

climate gis

Globalhealthy

mcp

80/100Currency A

geolens ↗

Self-hosted geospatial data catalog with semantic search (pgvector), OGC/STAC APIs, and map builder. Built on FastAPI, PostGIS, React, and MapLibre.

What's in it

Data offered docs-verified: Vector datasets from uploaded files, created datasets, or registered PostGIS tables · Feature-service snapshots from WFS, ArcGIS FeatureServer, or OGC API Features (one-shot copies) · Remote STAC assets as live references · Catalog metadata with fuzzy and optional semantic search
Formats docs-verified: Shapefile · GeoPackage · GeoJSON · GeoParquet · FlatGeobuf · KML · KMZ · zipped File Geodatabase · CSV · XLSX
Access docs-verified: Free and open source under Apache-2.0, self-hosted via Docker Compose; API calls use a JWT bearer token from admin login

Useful for

Use cases llm-inferred: Index shared-drive shapefiles and PostGIS tables for a flood-risk or drainage project data room, searchable by metadata · Publish a shared map of survey or site-investigation layers as a link or embed for client review · Serve catalogued vector layers to QGIS or ArcGIS via OGC API Features for FRA or environmental baseline work · Run buffer, clip, and intersect analysis on consultancy layers, such as constraint buffers around watercourses
Audience llm-inferred: GIS analysts · flood consultants · drainage engineers · ecologists · planners

Limitations

Constraints docs-verified: Read-only MCP server for coding agents (per docs) · Self-hosted: requires a running GeoLens stack of six containers (about 1.3 GB resident) · WFS, ArcGIS, and OGC API Features imports are one-shot copies, not live · Map previews capped at 500 features; dissolve is materialize-only · GeoLens's own REST API can change between minor releases; standards APIs track their specifications · Semantic search needs an embedding provider and admin setup · No MCP tool names are given in the metadata or docs

gis

Globalhealthy