geopera-mcp

The official Model Context Protocol server for the Geopera geospatial data platform. It exposes Geopera operations as MCP tools so an AI agent (Claude Desktop, Claude Code, or your own MCP client) can discover imagery, place and read back orders, run analytics, and more — with the same auth, scopes, and provenance as any other Geopera client.

Each tool is named after its operation id (e.g. orders.archive.place) and proxies the call to POST /v1/op/{operation_id}. The server imports nothing from the backend; it is a standalone client of the same typed surface the Python/TypeScript SDKs and CLI consume.

Install

pip install geopera-mcp

This installs the geopera-mcp console command, which speaks the stdio transport that MCP clients attach to. Requires Python 3.11+.

Related MCP server: Google Earth Engine MCP Server

Run

The server is configured through environment variables. At minimum:

export GEOPERA_API_URL="https://api.geopera.com"
export GEOPERA_API_TOKEN="gpra_..."   # a Geopera API key, or a session token
geopera-mcp

geopera-mcp runs over stdio by default — your MCP client launches it as a subprocess. Set MCP_TRANSPORT=http to serve the streamable-HTTP transport instead (for a hosted deployment).

Wire it into an MCP client

Add a geopera entry under mcpServers. For Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{
  "mcpServers": {
    "geopera": {
      "command": "geopera-mcp",
      "env": {
        "GEOPERA_API_URL": "https://api.geopera.com",
        "GEOPERA_API_TOKEN": "gpra_your_api_key"
      }
    }
  }
}

Restart the client and the Geopera tools appear. The agent's reach is exactly the scopes on the token — it can do nothing the token could not do directly.

Environment variables

Documentation

Full docs: docs.geopera.com/api-reference/sdks/mcp.

License

MIT