Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@netcdf-mcp What variables are in my NetCDF file?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that reads and understands NetCDF files, designed to give coding agents context about scientific data files.
uv pip install -e .Related MCP server: era5-mcp-server
Add to your MCP configuration file (~/.cursor/mcp.json):
{
"mcpServers": {
"netcdf-mcp": {
"command": "/path/to/uv",
"args": ["run", "--directory", "/path/to/netcdf-mcp", "netcdf-mcp"]
}
}
}Replace /path/to/uv with your uv binary location (e.g., /opt/homebrew/bin/uv) and /path/to/netcdf-mcp with where you cloned this repository.
netcdf-mcpTool | Description |
| Get dimensions, coordinates, variables, and attributes |
| List all data variables with shapes and units |
| Get file-level metadata |
| Detailed info about a specific variable |
| Statistical summary (min, max, mean, percentiles) |
| Extract sample values for inspection |
| Get coordinate values (time, lat, lon) |
| Check for NaN, outliers, and suspicious patterns |
| Compare two variables for compatibility and correlation |
| Get temporal extent in human-readable format |
| Get geographic extent (lat/lon bounds) |
| Verify CF conventions compliance |
| Human-readable explanation of a variable |
Check if a NetCDF file follows CF (Climate and Forecast) conventions.
Performs basic checks for common CF requirements.
Args: file_path: Path to the NetCDF file.
Returns: JSON string with compliance report.
Inputs: file_path (required). Returns: result
Check a variable for data quality issues: NaN/missing values, outliers, suspicious patterns.
Args: file_path: Path to the NetCDF file. variable_name: Name of the variable to check.
Returns: JSON string with data quality report.
Inputs: file_path (required), variable_name (required). Returns: result
Compare two variables for shape compatibility, range differences, and correlation.
Args: file_path: Path to the NetCDF file. variable1: Name of the first variable. variable2: Name of the second variable.
Returns: JSON string with comparison results.
Inputs: file_path (required), variable1 (required), variable2 (required). Returns: result
Generate a human-readable explanation of what a variable represents.
Uses standard_name, long_name, units, and other attributes.
Args: file_path: Path to the NetCDF file. variable_name: Name of the variable to explain.
Returns: JSON string with human-readable explanation.
Inputs: file_path (required), variable_name (required). Returns: result
Get values along a dimension (e.g., time steps, lat/lon coordinates).
Args: file_path: Path to the NetCDF file. dimension_name: Name of the dimension/coordinate to get values for. max_values: Maximum number of values to return (default: 100).
Returns: JSON string with dimension values.
Inputs: file_path (required), dimension_name (required), max_values. Returns: result
Get the complete structure of a NetCDF file.
Returns dimensions, coordinates, data variables, and global attributes.
Args: file_path: Path to the NetCDF file.
Returns: JSON string containing the file structure.
Inputs: file_path (required). Returns: result
Get all global attributes (metadata) from a NetCDF file.
Returns title, institution, history, conventions, and other file-level metadata.
Args: file_path: Path to the NetCDF file.
Returns: JSON string with all global attributes.
Inputs: file_path (required). Returns: result
Get the geographic extent (lat/lon bounds) of a NetCDF file.
Args: file_path: Path to the NetCDF file.
Returns: JSON string with spatial bounds.
Inputs: file_path (required). Returns: result
Get the time range of a NetCDF file in human-readable format.
Handles CF-compliant time coordinates with various calendars.
Args: file_path: Path to the NetCDF file.
Returns: JSON string with time range information.
Inputs: file_path (required). Returns: result
Get detailed information about a specific variable in a NetCDF file.
Returns dtype, shape, dimensions, all attributes, and fill value.
Args: file_path: Path to the NetCDF file. variable_name: Name of the variable to inspect.
Returns: JSON string with detailed variable information.
Inputs: file_path (required), variable_name (required). Returns: result
Extract a sample of actual data values from a variable for inspection.
Args: file_path: Path to the NetCDF file. variable_name: Name of the variable to sample. max_samples: Maximum number of samples to return (default: 100).
Returns: JSON string with sample values and their indices.
Inputs: file_path (required), variable_name (required), max_samples. Returns: result
Get statistical summary of a variable: min, max, mean, std, NaN count, percentiles.
Args: file_path: Path to the NetCDF file. variable_name: Name of the variable to analyze.
Returns: JSON string with statistical summary.
Inputs: file_path (required), variable_name (required). Returns: result
List all data variables in a NetCDF file with their shapes, dtypes, and units.
Args: file_path: Path to the NetCDF file.
Returns: JSON string with variable names, shapes, dtypes, and units.
Inputs: file_path (required). Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 13 tools
Most tools have clearly distinct inspection purposes, but get_variable_stats and check_data_quality overlap in their statistical/quality reporting, and get_file_structure subsumes list_variables and get_global_attributes. Overall, descriptions are clear enough to guide correct selection.
All tool names follow a consistent snake_case verb_noun pattern, using get_, list_, check_, compare_, and explain_ appropriately by action. There are no mixed naming conventions or unpredictable style shifts.
Thirteen tools is well within the ideal range for a focused NetCDF inspection server. Each tool covers a distinct aspect of file exploration without feeling redundant or bloated.
The server provides comprehensive read-only coverage for NetCDF files: structure, variables, attributes, dimensions, spatial/time bounds, statistics, samples, quality checks, comparisons, and CF compliance. There are no obvious dead ends for typical inspection workflows.
Read Physical AI datasets, projects, fleet and quality data. Requires authorized Avala access.
NOAA NCEI MCP — the national climate archive, via the keyless Access services
Build and run data workflows, analyses, reports and dashboards. Inspect and verify every step.
Read, query and check data against ODI standards hub ontologies (OWL, LinkML), and report gaps.