Skip to main content
Glama

elevation-mcp-server

by io.github.cyanheads
OverviewDiscussionsAdmin

Get Point Elevations

elevation_get_points
Read-onlyIdempotent

Look up ground elevation at up to 100 coordinates in one call, in meters and feet, with the dataset and resolution behind every point. Accepts up to 1,000 submitted positions; only the first 100 are validated and sampled, with counts and next_offset identifying the unvalidated remainder. Continue with originalPoints.slice(next_offset). Inside USGS 3DEP coverage (the US and its territories, plus much of Canada and Mexico) values come from 3DEP at 1–30 m resolution. Elsewhere they come from Open Topo Data: SRTM at about 30 m on land between 60°N and 56°S, and Mapzen terrain tiles beyond SRTM's coverage and over the ocean, where values below 0 m are sea-floor depths. A point with no data in any queried dataset returns status no_data instead of failing the call.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
pointsYes1–1,000 submitted positions as {lat, lon} objects in decimal degrees (WGS84), or one point object. Only the first 100 are validated and sampled; an invalid unserved tail is validated when resubmitted. Coordinate aliases latitude, longitude, lng, and long are accepted.
sourceNoElevation source: auto (default) uses USGS 3DEP where it has data and Open Topo Data (SRTM, with Mapzen terrain tiles where SRTM has no data) for the rest; usgs_3dep uses 3DEP only; opentopodata uses Open Topo Data only. Case is ignored and spaces or hyphens read as underscores, so USGS-3DEP and open topo data are accepted; 3dep, usgs, and epqs are also aliases for usgs_3dep.auto

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
errorNoPresent when the call failed. Absent on success.
noticeNoGuidance on points without data, Mapzen sea-floor values, and results mixing 3DEP with Open Topo Data.
pointsNoOne entry per processed position, in input order, at most 100.
attributionNoSources to credit for the returned values, one line per dataset that answered.
next_offsetNoZero-based index immediately after the processed prefix in this call, or null when exhausted. Continue with originalPoints.slice(next_offset), where originalPoints is the array submitted in this call; use each newly submitted array for its next slice.
source_modeNoThe source the call used (auto unless the caller chose one).
points_with_dataNoNumber of points with status ok.
processed_pointsNoPositions validated and sampled in this call, at most 100.
remaining_pointsNoUnprocessed, unvalidated positions remaining in this input.
requested_pointsNoSubmitted positions, including duplicates and coverage misses.
vertical_referenceNoHeight-reference context per answering dataset, including unverified references and whether heights share a datum. No vertical datum conversion is performed.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed13 schema fields changed
    • changedInput schema / properties / points / description
      Previous value: -"1–100 points as {lat, lon} objects in decimal degrees (WGS84). latitude, longitude, and lng keys are also accepted."New value: +"1–1,000 submitted positions as {lat, lon} objects in decimal degrees (WGS84), or one point object. Only the first 100 are validated and sampled; an invalid unserved tail is validated when resubmitted. Coordinate aliases latitude, longitude, lng, and long are accepted."
    • changedInput schema / properties / points / items / description
      Previous value: -"A point as {lat, lon} in decimal degrees (WGS84)."New value: +"A {lat, lon} point: latitude -90 to 90, longitude -180 to 180 (WGS84). The first 100 positions are validated; the unserved tail is unvalidated. latitude, longitude, lng, and long aliases are accepted."
    • removedInput schema / properties / points / items / properties
      Removed value: -{
-  "lat": {
-    "description": "Latitude in decimal degrees (WGS84), -90 to 90.",
-    "maximum": 90,
-    "minimum": -90,
-    "type": "number"
-  },
-  "lon": {
-    "description": "Longitude in decimal degrees (WGS84), -180 to 180.",
-    "maximum": 180,
-    "minimum": -180,
-    "type": "number"
-  }
-}
    • removedInput schema / properties / points / items / required
      Removed value: -[
-  "lat",
-  "lon"
-]
    • removedInput schema / properties / points / items / type
      Removed value: -"object"
    • changedInput schema / properties / points / maxItems
      Previous value: -100New value: +1000
    • changedOutput schema / anyOf
      Previous value: -[
-  {
-    "not": {
-      "required": [
-        "error"
-      ]
-    },
-    "required": [
-      "points",
-      "points_with_data",
-      "source_mode",
-      "attribution"
-    ]
-  },
-  {
-    "required": [
-      "error"
-    ]
-  }
-]New value: +[
+  {
+    "not": {
+      "required": [
+        "error"
+      ]
+    },
+    "required": [
+      "points",
+      "requested_points",
+      "processed_points",
+      "remaining_points",
+      "next_offset",
+      "points_with_data",
+      "source_mode",
+      "vertical_reference",
+      "attribution"
+    ]
+  },
+  {
+    "required": [
+      "error"
+    ]
+  }
+]
    • addedOutput schema / properties / next_offset
      Added value: +{
+  "anyOf": [
+    {
+      "maximum": 9007199254740991,
+      "minimum": -9007199254740991,
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "Zero-based index immediately after the processed prefix in this call, or null when exhausted. Continue with originalPoints.slice(next_offset), where originalPoints is the array submitted in this call; use each newly submitted array for its next slice."
+}
    • changedOutput schema / properties / points / description
      Previous value: -"One entry per input point, in input order."New value: +"One entry per processed position, in input order, at most 100."
    • addedOutput schema / properties / processed_points
      Added value: +{
+  "description": "Positions validated and sampled in this call, at most 100.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}
    • addedOutput schema / properties / remaining_points
      Added value: +{
+  "description": "Unprocessed, unvalidated positions remaining in this input.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}
    • addedOutput schema / properties / requested_points
      Added value: +{
+  "description": "Submitted positions, including duplicates and coverage misses.",
+  "maximum": 9007199254740991,
+  "minimum": -9007199254740991,
+  "type": "integer"
+}
    • addedOutput schema / properties / vertical_reference
      Added value: +{
+  "description": "Height-reference context per answering dataset, including unverified references and whether heights share a datum. No vertical datum conversion is performed.",
+  "type": "string"
+}
  2. First observed

TDQS

A4.1/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/openWorld/idempotent, yet the description adds substantial behavior beyond them: the hard truncation at 100 sampled points, the next_offset/counts continuation contract, and the fact that a data-less point returns status no_data rather than failing the call. It also discloses dataset-specific fallback rules and that sub-zero values over ocean are sea-floor depths.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The core purpose and the batching limit are front-loaded in the first sentence, and each subsequent sentence carries distinct information (continuation, coverage, source fallback, no_data). It is dense and runs long, but no sentence is filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with non-obvious coverage rules, source fallbacks, and a truncation contract, the description covers everything the agent needs to call it correctly and interpret results; the output schema carries the return shape, so nothing material is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% – both 'points' and the 'source' enum are fully documented in the schema, including aliases and the 1,000/100 validation split. The description largely restates the batching behavior and says nothing about parameter formats beyond what the schema already provides, so the baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb+resource: 'Look up ground elevation at up to 100 coordinates in one call', and adds scope (meters/feet, per-point dataset and resolution). This is clear, but it never names or contrasts with the sibling tools (elevation_get_grid, elevation_get_profile, elevation_check_line_of_sight), so the agent must infer the point-sampling distinction on its own.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete operational context: up to 1,000 positions may be submitted while only the first 100 are sampled, and it tells the agent to continue via originalPoints.slice(next_offset). It also explains when each source applies (3DEP coverage vs Open Topo Data/SRTM vs Mapzen). It stops short of saying when to prefer this tool over its siblings.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All elevation-mcp-server toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources