mcpField
S
swatgenx

swatgenx

Remote
swatgenx.comScience & Education1Updated 4w ago

About swatgenx

swatgenx is a remote MCP server for exploring and ordering SWAT+ watershed models across the lower 48 states. It also provides documentation, example-model results, and groundwater and PFAS monitoring data to environmental researchers and modelers.

Key features

  1. 01Search SWAT+ example models and calibration results
  2. 02Query groundwater well and PFAS inventories
  3. 03Preview watershed model contents
  4. 04Order, track, and download SWAT+ builds
  5. 05Search and read documentation

Use cases

  1. 01Compare example models and validation results
  2. 02Check groundwater wells or PFAS monitoring
  3. 03Build a SWAT+ model for a watershed

MCP support

Exposes SWATGenX documentation, data queries, model exploration, and model-ordering tools to MCP clients through remote streamable HTTP. Calls require SWATGenX sign-in or an API key.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://www.swatgenx.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

SWATGenX MCP Server

The agent-native front door to SWATGenX: SWAT+ watershed models for any river in the lower 48 states, built in the cloud from national datasets, with calibration as a separate step you run when you are ready.

Add it to Claude or any MCP client, and an agent can search the example-model catalog with its calibration and held-out validation scores, query a national groundwater well inventory and a national PFAS monitoring inventory, preview what a model of a watershed would contain, read the documentation, and order, track and download a real SWAT+ build under the same allocations as the web app.

Connect

claude mcp add --transport http swatgenx https://www.swatgenx.com/mcp

or, in a client's JSON config:

{ "mcpServers": { "swatgenx": { "type": "http", "url": "https://www.swatgenx.com/mcp" } } }

Every call needs a SWATGenX sign-in or an API key. Clients that support MCP authorization sign in with OAuth 2.1 through swatgenx.com; clients that set headers can send Authorization: Bearer <your API key>.

Tools

From 15 October 2026, the model, data and order tools are part of the paid plans; the documentation tools stay open to any account.

ToolTitleAccess from 15 Oct 2026
search_docsSearch the SWATGenX documentationany account
get_docRead a SWATGenX documentation chapterany account
get_engine_infoGet SWAT+ engine source, build guide and toolchainany account
get_access_infoGet access tier and allocationsany account
search_swat_modelsSearch SWAT+ example modelspaid plan
get_model_calibrationGet model calibration resultspaid plan
query_groundwaterQuery national groundwater wellspaid plan
query_pfasQuery PFAS monitoring (US or worldwide)paid plan
request_modelPreview a watershed model buildpaid plan
order_modelOrder a SWAT+ watershed modelpaid plan
get_order_statusGet build order statuspaid plan
list_my_modelsList your model orderspaid plan
download_modelGet a model download linkpaid plan

About

SWATGenX builds SWAT+ models for a USGS gauge, a HUC12 outlet or a HUC8 basin from NHDPlus HR hydrography, gSSURGO soils, NLCD land cover and PRISM weather, with an optional steady-state MODFLOW 6 groundwater model. Calibration against USGS streamflow is a separate step you run when you are ready. Its engine runs SWAT+ and MODFLOW 6 two-way, daily.

The national groundwater inventory is described in doi:10.5194/essd-2026-527.

The server implementation lives in the main SWATGenX repository; this repository holds the registry manifest (server.json) and the publish workflow.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.