Agentic MIKE+

A headless, natural-language-driven, automated modelling workflow for MIKE+.
Skills + an MCP server for Claude Code, Codex, Hermes, or OpenClaw: describe a goal in plain language and the agent inspects, edits, runs, reads, and plots a MIKE+ model. Automated modelling and analysis, end to end, without ever opening the GUI.

[!TIP]
Need a model to run? SWMMCanada builds one anywhere in Canada.
It is the upstream, open-data model builder for this agentic workflow: draw an area on a map and it assembles a ready-to-run stormwater model from Canadian open data (real municipal storm networks for 8 cities, synthesized everywhere else), giving the agent a real network to run, edit, read, and plot. Try the hosted demo with no install at swmm.h2ox.me, or see the project at h2ox.me.

Experimental / pre-release. One MCP server + skills wrapping DHI's Python stack (mikeplus / mikeio / mikeio1d). Verified end to end on the MIKE+ 2026 Sirius_RTC example. Sibling of agentic-swmm-workflow.

Install: just tell your agent

It is the AI era. You don't wire this up by hand. Paste this to your AI coding agent (Claude Code, Codex, Hermes, OpenClaw):

Install "Agentic MIKE+" for me: an MCP server + skills to drive MIKE+ headless.

1. Clone https://github.com/Zhonghao1995/Agentic-MIKE-Plus and skim its README.
2. With Python 3.11 x64 (mikeplus needs 3.9-3.11, not 3.12+):
     py -3.11 -m venv .venv
     .venv\Scripts\python.exe -m pip install -r requirements.lock
     .venv\Scripts\python.exe -m pip install -e ".[run]"
   (read/plot only, no license: drop the lock and use `pip install -e .`)
3. Register with me (Claude Code):
     claude mcp add mike-plus -- "<abs-repo>\.venv\Scripts\python.exe" -m mikeplus_mcp.server
   (Codex / Hermes / OpenClaw: copy config/mcp.sample.json)
4. Copy skills/* into ~/.claude/skills/, then run scripts/smoke_test.py (should find 10 tools).
5. Tell me the tools and which need a MIKE+ license (run/edit do; read/plot don't).

Needs Python 3.11 (x64). Two install profiles:

Read & plot. License-free and cross-platform: pip install -e . (no mikeplus).
Run & edit too. Windows + a licensed MIKE+ 2026: pip install -e ".[run]" (or pip install -r requirements.lock for the exact pinned environment).

mikeplus is an optional [run] extra, so teammates who only read results or make figures install nothing license-bound.

Why it matters

Natural-language-driven. Describe the task in plain words; the agent plans and runs it: no scripting, no GUI clicking. (A sub-agent did this autonomously.)
Fully headless. Runs with no GUI, on a workstation, a server, in CI, or under an agent. Built for batch and scenario automation.
MCP-native and portable. One server speaks the Model Context Protocol; works with Claude Code, Codex, Hermes, or OpenClaw via a single config line, and installs with pip.
Low barrier to share. Reading results and plotting need no MIKE+ license; only running or editing does. Teammates analyse model output with nothing but pip install.
Reproducible & tested. Deterministic tools, structured-JSON output, a pinned lockfile, and a license-free unit-test suite in CI, verified on a real model, not a chat-to-model black box.
Engine-agnostic and extensible. Results use a common schema (ready to sit beside SWMM and LSTM); add a tool or skill by dropping in a file.

Agentic MIKE+ overview: agentic-workflow advantages, core functionality and tools, and which capabilities need a MIKE+ license

How it works

Skills (markdown playbooks) tell the agent when and how; the agent calls MCP tools; each tool runs in an isolated worker subprocess that imports only mikeplus or mikeio*: the two cannot share a process. The server itself imports neither.

agent  ->  reads skills/*.md  ->  calls MCP tools  ->  workers (mikeplus / mikeio1d)

From one plain-language sentence to a finished MIKE+ run, automatically and without the GUI

Tools (one server, `mike-plus`)

Tool	Does	License
`mike_model_info`	model overview: simulations, scenarios, element counts	yes
`mike_get_values` / `mike_set_values`	read / change parameters (e.g. pipe diameter)	yes
`mike_run`	run a simulation headless, return `.res1d` + a parsed QA status (completed / errors / warnings)	yes
`mike_results_list` / `summary` / `read`	list contents / peaks / one time series	no
`mike_plot_rain_flow` / `timeseries` / `network`	stacked hydrograph / series / network map	no

Five skills (mike-model, mike-params, mike-runner, mike-results, mike-plot) orchestrate them.

Install the skills into any skills-aware agent (Claude Code, Codex, OpenCode, …) in one command, no clone needed:

npx skills add Zhonghao1995/Agentic-MIKE-Plus      # all 5; add --list to preview, or --skill <name> for one

Demo: `Sirius_RTC` (MIKE 1D, 568 nodes, 576 links)

Sirius_RTC network layout

Rainfall-runoff hydrograph for the busiest pipe

Full evidence (commands, outputs, and the honest license boundary) is in docs/verification.md.

Development

The engine-agnostic core is covered by a license-free test suite (no MIKE+ / mikeplus needed) that also runs in CI:

pip install -e .            # read/plot core (add ".[run]" for run/edit)
pip install pytest
pytest                      # ~0.5 s, no license required

The tests pin the result schema, the res1d column matcher, the engine-log QA parser, and tool discovery, so a change can't silently break them. Add a tool or skill by dropping a file under mikeplus_mcp/tools/ or skills/ (auto-discovered), and ship a test with it.

License

MIT © 2026 Zhonghao Zhang, University of Victoria. Built on DHI's mikeplus / mikeio / mikeio1d and the Model Context Protocol. MIKE+ is a product of DHI; running models requires a valid DHI license.