KISS knowledge infrastructure skills
SkillCommunity
Per-model SKILL.md packages with tools, checks and failure fixes for EPANET, SWMM, HEC-RAS, HEC-HMS and MODFLOW 6. By lzwei196.
187 starsUpdated 2 days agoGitHub data from 5 Nov 2026
At a glance
- Type
- Skill
- Works with
- EPANET, EPA SWMM, HEC-RAS, HEC-HMS, MODFLOW / GMS
- Tool versions
- EPANET 2.2.0; SWMM 5.2 via pyswmm; HEC-RAS 6.7 Beta 5 steady solver under Wine; HEC-HMS 4.14 Linux build; MODFLOW 6 via FloPy
- Operating systems
- Not stated in its documentation
- Needs a EPANET licence
- No
- AI clients documented
- Not stated in its documentation
- Licence
- MIT
- Language
- Python
What it can do
- Each model package pairs a SKILL.md with stage scripts, a preflight check and a table of known failure symptoms, diagnoses and remedies.
- The EPANET package converts demands and network parameters, runs EPANET 2.2 and parses output.
- The HEC-HMS package runs the official 4.14 Linux engine headless through its Jython API and exports DSS results to CSV.
- The MODFLOW 6 package builds and runs FloPy models, including transport and streamflow routing stages.
- Part of a catalogue of more than 100 Earth-science model packages from a published study.
Limitations
- Research packages tested on the authors' sites; the HEC-RAS package validates only the steady solver, run under Wine.
- The HEC-HMS package also contains surrogate scripts that are not HEC-HMS; the skill marks them.
- The skills tell the agent to run preflight checks and use the real model, so each model must be installed locally.
Install
git clone https://github.com/lzwei196/KISS---Knowledge-Infrastructure-for-Scientific-Simulation.git
cd KISS---Knowledge-Infrastructure-for-Scientific-Simulation
pip install -e ki_tools_common/MCP servers and skills run code on your computer. Review the source before installing.
AI-generated models, drawings and results must be checked by a qualified engineer before use.