@cyanheads/elevation-mcp-server

Public Hosted Server: https://elevation.caseyjhand.com/mcp

Overview

Elevation and terrain analysis anywhere on Earth, from USGS 3DEP (1–30 m across the US) and Open Topo Data worldwide. Look up spot heights, measure a route's climbs and grades, find an area's high and low points, and check whether terrain blocks a line of sight. No API key required.

Tools

Agent skill

The elevation-guide covers route geometry, candidate high-point refinement, and visual or radio sightline assessment. Plugin installs discover it in root skills/; npm, Docker, and MCPB distributions include it. MCP clients can read skill://elevation-guide/SKILL.md as a resource or use the Skills extension.

Related MCP server: Geospatial Tools MCP Server

Capability reference

Every successful result includes vertical_reference on both response surfaces. Public SRTM uses EGM96; NAVD88 is typical for 3DEP in the conterminous US, but the returned raster reference is unverified and can vary elsewhere. Mapzen and self-hosted Open Topo Data references are unknown/variable. No-data-only results establish no height reference. Heights are returned without transformation; mixed datasets are not normalized to a common datum. water_surface_m needs a compatible terrain reference; zero or source: opentopodata alone does not establish compatibility.

elevation_get_points tool

• points: 1–1,000 {lat, lon} objects in decimal degrees (WGS84), or a single point. Only the first 100 are validated and sampled; the remainder stays unvalidated

• requested_points, processed_points, and remaining_points count submitted positions, including duplicates and misses. next_offset is the index after this call's processed prefix, or null when exhausted; continue with originalPoints.slice(next_offset), where originalPoints is the array submitted in that call. Use each newly submitted array for the next slice

• Each point comes back ok or no_data (a miss never fails the call), with elevation_m / elevation_ft, dataset, and resolution_m (omitted for mapzen); 3DEP answers add raster_id and acquisition_date. points_with_data counts the hits

elevation_get_profile tool

• path: 2–1,000 vertices in travel order (consecutive duplicates dropped); samples: 2–250 evenly spaced points along it, endpoints included (default 100)

• summary carries total_distance_m, ascent_m / descent_m (and feet), start, end, min, and max elevation, highest_point / lowest_point, and max_grade_pct / min_grade_pct with where each occurs; ascent depends on the reported sample_interval_m. Fails as degenerate_path (under 1 m of route), antipodal_path (add an intermediate waypoint between the identified zero-based input vertices), or no_coverage (no sample has data)

• include_samples (default true) returns samples[] (distance, position, elevation, grade, dataset, and resolution per sample); false omits it and the sample table, and the rest of the result is unchanged

elevation_get_grid tool

• south, west, north, east edges in decimal degrees (a box can't cross longitude 180); rows and cols 2–25 each (default 10), with rows × cols at most 250

• elevations_m and cell_datasets matrices indexed [row][col] (row 0 north, column 0 west, null without data), plus summary.highest, summary.lowest, mean_elevation_m, and relief_m. Fails as invalid_bbox, too_many_cells, or no_coverage

• Nodes are point samples with the edges included, not cell averages

elevation_check_line_of_sight tool

• observer and target points, at most 1,000 km apart; observer_height_m (default 1.7) and target_height_m (default 0), each 0–10,000 m above ground; earth_model flat, geometric, optical (default, κ 0.13), or radio (κ 0.25); samples 3–250 (default 100); optional water_surface_m (-500 to 9,000) and frequency_mhz (30–300,000)

• verdict is clear, blocked, or indeterminate (samples without data leave the line unconfirmed), with min_clearance_m / min_clearance_ft, the limiting_point, and the first_obstruction when blocked. Fails as same_endpoints (under 1 m apart), sightline_too_long (over 1,000 km apart), or endpoint_no_data

• frequency_mhz adds fresnel: a sufficient, insufficient, or indeterminate verdict against the 60% free-space bar of the first Fresnel zone, min_clearance_ratio (clearance over zone radius), and the sample that limits it, which is often not limiting_point

• Terrain only: buildings and vegetation count only as far as the elevation source captures them. Where Mapzen reports sea-floor depth, clearance is measured to the sea surface; USGS 3DEP values below 0 m (bay floor in some bays, or dry land) count as received unless water_surface_m sets a water level, which then applies to every sample

Failures shared by every tool

A coverage miss is never an error: that point has no data. Any provider failure fails the whole call with no partial result.

Features

Built on @cyanheads/mcp-ts-core: stdio and Streamable HTTP transports, pluggable auth (none / jwt / oauth), swappable storage (in-memory, filesystem, Supabase, Cloudflare KV/R2/D1), structured logging with optional OpenTelemetry tracing.

Elevation-specific:

• source on every tool: auto (default) queries USGS 3DEP inside its coverage and sends 3DEP misses and everything outside it to Open Topo Data; usgs_3dep and opentopodata pin one provider. Only a coverage miss falls back, never an outage

• Open Topo Data answers from SRTM GL1 v3 (about 30 m, land between 60°N and 56°S), then Mapzen terrain tiles where SRTM has no tile: high latitudes, Antarctica, and ocean bathymetry

• Per-call caps of 100 points or 250 samples or grid cells, inside a 45 s sampling budget; each 3DEP point is its own request (6 concurrent, 10 per second), while Open Topo Data takes 100 points per request

• The public Open Topo Data instance is paced under its published limits: one request at a time, starts at least 1.1 s apart, at most 1,000 in any trailing 24 hours per server process

• Profiles, grids, and sightlines are computed locally from point samples, with great-circle distances and resampling and earth curvature with standard refraction

Agent-friendly output:

• Provenance on every value: each point, profile sample, grid cell, and sightline point names its dataset (usgs_3dep, srtm30m, mapzen) and, except for Mapzen, its resolution_m (resolution_m_range on profiles and grids); computed results add datasets_used counts, and a Sources: line credits every dataset that answered

• Absent stays absent: a point or sample without data omits its elevation and a grid cell is null, never 0; summaries report how many values had data

• Notices flag what changes interpretation: results mixing 3DEP and Open Topo Data, Mapzen values below 0 m (sea-floor depths), 3DEP values below 0 m on a sightline, sample spacing coarser or finer than the source, thin clearance margins, and a clear radio path short of 60% Fresnel clearance

Getting started

Connect to the public instance:

claude mcp add --transport http elevation-mcp-server https://elevation.caseyjhand.com/mcp

Every hosted caller shares one Open Topo Data allowance of 1,000 requests a day. Under auto, 3DEP answers inside its coverage and Open Topo Data handles the rest. For sustained global use, run your own instance:

claude mcp add elevation-mcp-server -- npx -y @cyanheads/elevation-mcp-server

Other MCP clients use this local configuration:

{
  "mcpServers": {
    "elevation-mcp-server": {
      "command": "npx",
      "args": ["-y", "@cyanheads/elevation-mcp-server"]
    }
  }
}

See Running the server for source and Docker setup.

Configuration

Any OPENTOPODATA_BASE_URL other than the public instance is treated as self-hosted (Open Topo Data is MIT-licensed and runs in Docker): 4 concurrent requests, no rate windows. It must serve datasets named srtm30m and mapzen and accept at least 100 locations per request, or calls fail with opentopodata_config_rejected.

See .env.example for every server setting and the common framework overrides.

Known limitations

• Open Topo Data's public limits. The public instance allows 100 locations per request, 1 request per second, and 1,000 requests per day per IP address. Each server process paces itself under them, but its count is process-local and resets on restart. Several processes on one machine (a stdio server per client session), or other clients on the same address, also draw on the upstream's count, so the upstream can refuse first (opentopodata_rate_limited).

• Shared deployments. A deployment serving many users sends all their requests from one address, so on the public instance they share one 1,000-a-day allowance: about 1,000 point lookups, or 333 full 250-sample calls, and as few as 111 when retries spend requests too. Once it is spent, every call that needs Open Topo Data fails with opentopodata_daily_limit for up to 24 hours, including an auto call with a single 3DEP miss, so global answers on such a deployment are best-effort; pointing OPENTOPODATA_BASE_URL at its own instance (loaded with srtm30m and mapzen) removes the cap. The server has no caller identity to ration by, so one caller's heavy calls slow USGS 3DEP and can spend the public day for everyone.

• Global resolution. SRTM is a 30 m radar surface model. Mapzen's 1 arc-second grid is interpolated in places from coarser sources (GMTED2010 at 7.5 arc-seconds over parts of the high latitudes, ETOPO1 at about 1.8 km over the open ocean). Outside 3DEP coverage, profile ascent is underestimated and line of sight misses terrain narrower than the source's true resolution.

• Whole-meter values. Open Topo Data's datasets are integer rasters, and the upstream rounds the interpolated result to the nearest meter, so ascent and descent over gentle terrain include 1 m quantization steps.

• Water. SRTM reports water inside a land tile as 0 m. Beyond SRTM's tiles, Mapzen reports sea-floor depth, and points, profiles, and grids include it. 3DEP values over water depend on the raster: a hydro-flattened surface, sea level, or bathymetry. Line of sight measures clearance to 0 m over Mapzen values below 0, but uses 3DEP values below 0 m as received, since nothing in a USGS answer tells bay floor from dry land below sea level; its notice flags them, and water_surface_m measures a line over water to the water instead.

• Mixed models. 3DEP is a bare-earth DEM, typically NAVD88 in the conterminous US; the exact returned raster reference is unverified and can vary elsewhere. Public SRTM is a radar surface model that partly includes canopy and buildings, using EGM96. Mapzen blends regional sources; self-hosted dataset names do not verify the reference. Heights are not transformed to a common datum, and datum alone does not explain differences between providers.

• Sampling. Features narrower than the sample spacing are missed: a ridge between line-of-sight samples, a summit between grid nodes (re-grid a smaller box around summary.highest to refine it), short climbs between profile samples. Spacing is always reported.

• USGS 3DEP throughput. The point query service has no published limit and no batch endpoint: about 10 points per second at this server's pacing, hence the 250-sample cap and 45 s budget. A call whose lookups would push the queue past a running call's budget is refused at once with sampling_deadline_exceeded and data.retryAfter, so two 250-sample calls at once run in turn: the second can be retried after about 25 s.

• Acquisition dates from USGS are passed through raw in M/D/YYYY form, which sometimes carries a zero month or day; any other form is omitted.

• Antimeridian. A grid box cannot cross longitude 180, so split it into two calls. Paths and sightlines crossing it work.

• Line of sight is limited to 1,000 km, where the parabolic curvature approximation overstates the midpoint rise by up to about 10 m. Its Fresnel check covers the first zone against the 60% bar at the chosen earth model, at the sample positions only, with no diffraction loss or worst-case k factor.

Running the server

Requires Bun 1.4.0+ or Node.js 24+. Both elevation providers are keyless. For heavy global use, set OPENTOPODATA_BASE_URL to a self-hosted Open Topo Data instance with srtm30m, mapzen, and at least 100 locations per request.

git clone https://github.com/cyanheads/elevation-mcp-server.git
cd elevation-mcp-server
bun install
cp .env.example .env

For Docker over stdio:

docker run -i --rm -e MCP_TRANSPORT_TYPE=stdio ghcr.io/cyanheads/elevation-mcp-server:latest

For HTTP from source:

MCP_TRANSPORT_TYPE=http MCP_HTTP_PORT=3010 bun run start:http
# http://localhost:3010/mcp

Local development

• Build and run the production version:

  # One-time build
  bun run rebuild
  
  # Run the built server
  bun run start:http
  # or
  bun run start:stdio

• Run checks and tests:

  bun run devcheck  # Lints, formats, type-checks, and more
  bun run test      # Runs the test suite

Project structure

Development guide

See CLAUDE.md for development guidelines and architectural rules. The short version:

• Handlers throw, framework catches — no try/catch in tool logic

• Use ctx.log for logging, ctx.state for storage

• Register new tools in allToolDefinitions in src/mcp-server/tools/definitions/index.ts

• Wrap external API calls: validate raw → normalize to domain type → return output schema; never fabricate missing fields

Data sources and attribution

Every successful result's Sources: line credits each dataset that answered, and a response that used any Mapzen value carries the full Mapzen attribution there. The public Open Topo Data instance publishes no terms beyond its usage limits; its server software is MIT-licensed and self-hostable.

Mapzen terrain tiles attribution, verbatim from the tilezen/joerd attribution document:

• ArcticDEM terrain data DEM(s) were created from DigitalGlobe, Inc., imagery and funded under National Science Foundation awards 1043681, 1559691, and 1542736;

• Australia terrain data © Commonwealth of Australia (Geoscience Australia) 2017;

• Austria terrain data © offene Daten Österreichs – Digitales Geländemodell (DGM) Österreich;

• Canada terrain data contains information licensed under the Open Government Licence – Canada;

• Europe terrain data produced using Copernicus data and information funded by the European Union - EU-DEM layers;

• Global ETOPO1 terrain data U.S. National Oceanic and Atmospheric Administration

• Mexico terrain data source: INEGI, Continental relief, 2016;

• New Zealand terrain data Copyright 2011 Crown copyright (c) Land Information New Zealand and the New Zealand Government (All rights reserved);

• Norway terrain data © Kartverket;

• United Kingdom terrain data © Environment Agency copyright and/or database right 2015. All rights reserved;

• United States 3DEP (formerly NED) and global GMTED2010 and SRTM terrain data courtesy of the U.S. Geological Survey.

This server is independent of the U.S. Geological Survey and of Open Topo Data, and is not endorsed by either.

Contributing

Issues are welcome at the issue tracker.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details.