← Directory
mcp · Global
epanet-mcp-server
Model Context Protocol (MCP) Server for loading water network models, running simulations, modifying parameters, and generating complex what-if scenarios through natural language
Claim this listing
Represent epanet-mcp-server? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.
What's in it
- Data offered docs-verified
- water distribution network model loading and inspection · hydraulic and quality simulation · parameter modification of network models · what-if scenario generation (demand, leakage, contamination, pressure, pump, valve, multi-failure)
- Coverage llm-inferred
- Global (per ENTRY regions); applies to whichever EPANET network the user loads
- Resolution llm-inferred
- Set per network model via set_hydraulic_timestep and set_quality_timestep; no default stated in DOCS
- Formats docs-verified
- MCP stdio · MCP SSE · EPANET .inp network files
- Access llm-inferred
- Free and open source under MIT licence; local install via pip install -e .; no authentication; requires Python ≥ 3.10, ePyT ≥ 2.0 and mcp ≥ 1.0
Useful for
- Use cases llm-inferred
- Peak-hour pressure check on a water distribution model to assess the impact of a new development connection · Reservoir head or pump schedule what-if testing to confirm service pressures under changed supply conditions · Burst and leakage scenario screening on a supply network model · Chlorine or contaminant injection screening for water quality risk assessments
- Audience llm-inferred
- water distribution network modellers · hydraulic engineers · water utility and network planners · water quality engineers
Limitations
- Constraints llm-inferred
- Requires a local Python install with ePyT and mcp; DOCS describe no hosted endpoint · No authentication (ENTRY auth is null) · Bundled networks are standard benchmark models, not UK network data; the user must supply their own models for UK work · Tool argument schemas are not documented in DOCS · ENTRY lists no tools in metadata, so the tool list relies on DOCS
- Licence
- MIT
- Authentication
- Unknown — review source docs
- GitHub stars
- 9
- Last commit
- 02/06/2026
- Last successful page check
- 14/10/2026
- Page health
- healthy
- Package
- pypi:epyt
- Verification
- Not independently verified
Auto-extracted, unverified fields: tools, install, auth. Check upstream documentation.
Provenance & evidence
Tools/install/auth auto-extracted from README; quality score heuristic (stars, recency, tool count, licence).
Metadata observed 15/10/2026 · research sweep: github-search