Hydrata MCP Server¶

Hydrata provides a native Model Context Protocol (MCP) server for AI coding assistants. Manage flood model inputs, run simulations, and retrieve results directly from your AI assistant.

Endpoint: https://hydrata.com/mcp/ Transport: StreamableHTTP Registry: registry.modelcontextprotocol.io

Available Tools¶

The MCP server exposes 9 tools for the complete flood modeling workflow:

Tool	Description
`list_projects`	List all projects for the authenticated user
`create_project`	Create a new flood modeling project
`get_project`	Get project details including scenarios and runs
`create_scenario`	Configure model inputs — DEM, boundaries, friction, rainfall
`list_scenarios`	List scenarios within a project
`run_simulation`	Submit an ANUGA simulation to cloud compute
`get_run_status`	Check simulation progress and status
`get_results`	Retrieve simulation outputs — depth maps, velocity, extents
`list_runs`	List all runs for a scenario

Supported Clients¶

Claude Code — Anthropic's CLI assistant
GitHub Copilot — VS Code with MCP support
Cursor — AI-first code editor

Any MCP-compatible client can connect via the StreamableHTTP transport.

Authentication¶

All MCP requests require a Bearer token. Get your API key by:

Sign up for a free Hydrata account
API key is provisioned on email confirmation
Pass it as Authorization: Bearer <your-token> header

Quick Start¶

# Add Hydrata MCP server to Claude Code
claude mcp add hydrata https://hydrata.com/mcp/ \
  --transport streamable-http \
  --header "Authorization: Bearer $HYDRATA_TOKEN"

Then ask Claude to create a flood model:

> Create a flood model for my site using a 5m DEM

See Getting Started for a complete walkthrough.