# GeoLens [Health: Active]

**Category:** 🗺️ Location Services  
**Repository:** https://github.com/geolens-io/geolens  
**GitHub Stars:** 278  
**Views:** 2  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/geolens

## Description
Read-only access to a self-hosted GeoLens spatial catalog: datasets, features, maps, sandboxed SQL.

## Claude Desktop Quick Installation
No confirmed install details yet, so no config is published. Follow the setup steps in the repository above.

## Documentation

## What GeoLens MCP server does

GeoLens MCP server gives AI agents read-only access to a GeoLens spatial catalog. The catalog brings together datasets from file storage, PostGIS tables, ArcGIS and WFS services, OGC API Features services, and remote STAC assets. Each dataset can include metadata, provenance, source information, freshness status, geometry details, coordinate reference information, feature counts, and version history.

The MCP interface is intended for agents that need to find and inspect geographic data rather than administer the catalog. It exposes catalog content alongside the maps built in GeoLens, allowing an agent to work with the same spatial resources available through GeoLens’s open APIs.

## How it works

An agent sends read requests through MCP to the GeoLens deployment. The server reads from the connected catalog and returns information about available datasets, map resources, and geographic features. Queries can also be expressed as sandboxed SQL, which supports data exploration while constraining the operation to a read-only execution context.

GeoLens itself runs on infrastructure controlled by the deploying team. Its catalog can reference data held in external systems, including existing PostGIS databases and remote STAC assets, as well as data uploaded into GeoLens. This makes the MCP layer useful when an agent needs catalog-aware access instead of direct connections to each underlying source.

## Setup and configuration

GeoLens is designed for self-hosting with Docker Compose. Deploy the GeoLens application and connect it to the spatial data sources that should appear in the catalog. The MCP server is then used against that self-hosted catalog, keeping access tied to the organization’s GeoLens instance rather than a shared hosted service.

The project also exposes OGC API and STAC interfaces for software such as QGIS, scripts, and other geospatial clients. Those interfaces complement MCP access but are separate ways to consume the same catalog and data resources.

## Tools and capabilities

- Browse and inspect catalog datasets.
- Retrieve geographic features from the catalog.
- Access maps managed in GeoLens.
- Execute sandboxed SQL for read-only spatial queries.
- Work with dataset metadata and recorded provenance.
- Query data that may originate from files, PostGIS, web services, or STAC catalogs.

## Limitations and notes

Access through this server is read-only. It is intended for querying and inspection, not for uploading datasets, editing metadata, changing maps, managing users, or restoring dataset versions.

SQL execution is sandboxed, so it should be treated as a constrained query facility rather than unrestricted database access. The available results depend on the datasets and sources configured in the connected GeoLens deployment.

GeoLens can also serve imagery and other spatial resources through its catalog and open APIs, but the MCP interface described here is centered on catalog reads, features, maps, and sandboxed SQL.

_Full upstream README: https://allmcps.com/mcp/geolens/readme_