Skip to content

JordanGunn/gdal-mcp

gdal-mcp

MCP server exposing GDAL/Rasterio operations to AI agents, with a reflection middleware that requires structured justification before executing operations whose methodology matters (CRS choice, resampling method, query extent).

CI License: MIT Python 3.11+ FastMCP 2.0 PyPI Downloads

Install

Via uvx (recommended)

uvx --from gdal-mcp gdal --transport stdio

Via Docker

docker build -t gdal-mcp .
docker run -i gdal-mcp gdal --transport stdio

Local development

git clone https://github.com/Wayfinder-Foundry/gdal-mcp.git
cd gdal-mcp
uv sync
uv run gdal --transport stdio

Configure your MCP client

Claude Desktop

Add to claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\, Linux: ~/.config/Claude/):

{
  "mcpServers": {
    "gdal-mcp": {
      "command": "uvx",
      "args": ["--from", "gdal-mcp", "gdal", "--transport", "stdio"],
      "env": {
        "GDAL_MCP_WORKSPACES": "/path/to/your/geospatial/data"
      }
    }
  }
}

Restart Claude Desktop. The MCP server indicator should appear, and the raster_* and vector_* tools become available.

Workspace scoping

GDAL_MCP_WORKSPACES is a colon-separated list of directories the server is allowed to touch. If unset, all paths are allowed and a warning is logged.

Optional tool-surface flags: RASTER=true, VECTOR=true. See docs/ENVIRONMENT_VARIABLES.md for the full set.

Tools

  • Raster: raster_info, raster_convert, raster_reproject, raster_stats, raster_query
  • Vector: vector_info, vector_convert, vector_reproject, vector_clip, vector_buffer, vector_simplify, vector_query
  • Resources: catalog (workspace://...), metadata (metadata://...), reference (reference://...), query results (query://result/{id})
  • Prompts: justify_crs_selection, justify_resampling_method, justify_query_extent (and more under src/prompts/)

See TOOLS.md for parameters, return shapes, and worked examples.

The reflection middleware

Tools whose methodology matters refuse to execute until the calling agent produces a structured justification. The flow is:

  1. Agent calls e.g. raster_reproject(dst_crs="EPSG:3857", resampling="cubic", ...).
  2. Middleware checks .preflight/justifications/{domain}/ for a matching hash.
  3. On miss, the call raises ToolError with a hint pointing at the relevant prompt (e.g. justify_crs_selection).
  4. Agent calls the prompt, fills out the Justification schema (intent, alternatives considered, choice, tradeoffs, confidence), and re-invokes the tool with a __reflection payload.
  5. The justification is cached domain-keyed, so a CRS rationale for EPSG:3857 satisfies both raster_reproject and vector_reproject on subsequent calls.

See docs/REFLECTION.md for the schema and cache layout, and docs/PHILOSOPHY.md for why this exists.

Documentation

Troubleshooting

Access denied: path outside allowed workspaces — set GDAL_MCP_WORKSPACES to include the directory in question (see "Workspace scoping").

MCP client doesn't see the server — verify uvx --from gdal-mcp gdal --help runs on its own, then restart the client after editing its config file.

License

MIT — see LICENSE.

Built on FastMCP, Rasterio, pyogrio, and Shapely.

About

Model Context Protocol server that packages GDAL-style geospatial workflows through Python-native libraries (Rasterio, GeoPandas, PyProj, etc.) to give AI agents catalog discovery, metadata intelligence, and raster/vector processing with built-in reasoning guidance and reference resources.

Topics

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Packages

 
 
 

Contributors