USDA NASS MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| NASS_ENV | No | A flag for telling the MCP server to skip local dev steps | |
| transport | No | Transport protocol for the MCP server (streamable-http or stdio) | streamable-http |
| NASS_API_KEY | Yes | Your NASS API key (required) | |
| NASS_MCP_HOST | No | The host to run the MCP server on | 0.0.0.0 |
| NASS_MCP_PORT | No | The port to run the MCP server on | 8000 |
| NASS_MCP_FORMAT | No | The format to return data (CSV, JSON, or XML) | CSV |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_full_datasetC | Get all available data for a given commodity, location, and time. |
| get_db_record_countB | Get the number, or count, of database records of a given commodity for a location and time. |
| get_param_valuesB | Get all possible values of a query parameter by its name. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| get_parameter_names | List of all possible query parameter names |
| get_operators | List of operators that can be appended to parameter names in a query |
TDQS
Scored across 3 tools
The three tools have clearly distinct purposes: retrieving full data, retrieving only a record count, and discovering parameter values. An agent can easily tell which tool to use for each task.
All tool names follow a consistent snake_case get_<object> pattern. The naming is predictable and readable across the entire small tool set.
Three tools are well within the appropriate range for a focused data-query server, and each tool serves a distinct necessary function. There is no obvious redundancy or padding.
The surface covers core data retrieval, counting, and parameter-value discovery, but lacks an explicit way to discover parameter names or datasets before querying. This is a minor gap that a knowledgeable agent can work around.