Agentic SWMM is an automated, auditable, and memory-informed framework for reproducible stormwater modelling, integrating QGIS and EPA SWMM through the aiswmm runtime, reusable Skills, and MCP interfaces, with QA verification, provenance tracking, calibration support, and Codex, Hermes, Claude code as well as OpenClaw compatibility

What's in it

Data offered docs-verified: SWMM input files (model.inp) · SWMM report and binary outputs (.rpt, .out) · run manifests, command traces and QA summaries · parsed peak-flow metrics · rainfall-runoff figures · calibration summaries · fuzzy uncertainty summaries · audit records (experiment_provenance.json, comparison.json, experiment_note.md) · modelling memory summaries · SWMM networks from SWMMCanada (Canada) or SWMManywhere synthesis (elsewhere)
Coverage docs-verified: Canada: real storm networks for 35 cities, synthesis elsewhere in Canada; SWMManywhere-based synthesis outside Canada
Formats docs-verified: SWMM INP · SWMM RPT · SWMM OUT (binary) · JSON · GeoJSON · Parquet · PNG · Word (.docx) · Markdown (Obsidian-ready notes)
Access docs-verified: Free, open-source under MIT licence; local install via one-line installer, Docker image or pip (aiswmm); LLM provider API key needed for most routes, three of ten need none

Useful for

Use cases llm-inferred: Stormwater drainage model build from GIS inputs with QA and audit trail · Peak-flow and rainfall-runoff screening for drainage strategy options · Reproducible SWMM run handover with provenance records for client review
Audience llm-inferred: drainage engineers · hydrologists · stormwater modellers · GIS analysts

Limitations

Constraints llm-inferred: Pre-1.0 (v0.9.4), so interfaces may change · Real storm networks only for 35 Canadian cities via SWMMCanada; elsewhere SWMManywhere synthesis, which is not a calibrated or validated network · Validation rows are evidence-bounded: GIS preprocessing concept, prepared-input path and uncertainty smoke are not calibrated performance claims · Raw GIS path handles structured layers only, not arbitrary CAD/GIS recognition · ENTRY lists region as United States, but DOCS do not describe US or UK coverage, so UK suitability is unverified · Agent features need an LLM provider; one-line installers run a remote script · ENTRY auth says no credentials documented, but DOCS describe API key configuration · No MCP tool names or transport documented in DOCS; ENTRY tools list is empty · One evidence row (v0.7.1) states calibration is next-milestone scope

hydrology gis flood

United Stateshealthy

mcp

74/100Currency A

local-austin-mcp ↗

Free, open-source MCP server (Apache-2.0) giving Claude plain-English access to official Austin & Travis County data — property records, permits, flood zones, city code text, schools, plus active MLS listings. Built by Neuhaus Realty Group.

What's in it

Data offered docs-verified: active MLS for-sale listings (active and under-contract only) · neighbourhood lookup · Travis, Williamson and Hays CAD property records · county tax bills, exemptions and delinquencies · MUD/PID special-district overlay · FEMA NFHL flood zones · City of Austin permits · code-compliance cases · zoning, lot dimensions and plats · municipal code text (Austin, plus Leander, Round Rock, Dripping Springs) · tree permits · 311 requests · APD crime, arrests, use-of-force and dispatch data · AFD dispatches · council voting records · city budget · council district, school district, ESD and voter precinct lookup · TEA school ratings and AISD attendance zones · library branches · parks · fire stations · restaurant inspection scores · roadway work zones · Animal Center intakes and outcomes · TxDOT highway projects · trash/recycling collection days · CapMetro real-time bus and rail · NWS alerts · Lake Travis and Highland Lakes level and 30-day trend · drive time and distance between addresses
Coverage docs-verified: Austin, Texas metro: City of Austin plus Travis, Williamson and Hays counties
Update frequency docs-verified: Live, per request: each tool queries its upstream source at call time; no fixed update cadence is documented
Formats docs-verified: MCP stdio
Access docs-verified: Free and open source under Apache-2.0; no API key or login required. Optional AUSTIN_SODA_APP_TOKEN (Socrata app token) raises the rate limit on data.austintexas.gov and datahub.austintexas.gov.

Useful for

Use cases llm-inferred: Flood zone, permit and code-case screening for a site due-diligence note · Zoning, lot dimensions and municipal code text lookup for a pre-application brief · Tree-permit and roadway work-zone check before site clearance or access planning · Special-district (MUD/PID) and appraisal context for a development or drainage-authority brief
Audience llm-inferred: flood consultants · planners · civils engineers · GIS analysts

Limitations

Constraints llm-inferred: Austin, Texas and Travis, Williamson and Hays counties only; no UK data, so relevance to a UK consultancy is low, although the catalogue lists regions as Global · Early-stage project run by a single maintainer, so expect gaps and possible breaking changes · MLS data covers active and under-contract listings only, with no sold comps · Listings, neighbourhood and blog content depend on neuhausre.com and the Neuhaus VOW API · Socrata rate limits apply unless an app token is set · Commute tool uses demo routing servers with no real-time traffic · Catalogue auth field says an API key is required, but docs say no key is needed · Catalogue tool list does not match docs: docs state 43 tools and list austin_utility_providers, austin_relocation, austin_local_voices and austin_health, which are absent from the metadata, while the metadata lists health_check

flood planning civils-bim

Some details auto-extracted · unverified

Globalhealthy

mcp

72/100Currency A

mcp-server ↗

MCP server for Polish real estate: 8M+ transaction prices from notarial deeds (RCN registry, not listings), plus per-parcel context by cadastral ID - zoning, flood risk, heritage, building permits, agricultural land and transit

flood gis planning environment

Some details auto-extracted · unverified

Europehealthy

mcp

71/100Currency A

RAS Commander MCP ↗

MCP Server to Access HEC-RAS Project Data via Claude Desktop, Claude Code and other LLMs

What's in it

Data offered docs-verified: server/version information · project length units · plan descriptions · selected project/plan metadata · bounded read-only HEC-RAS project and plan text
Formats docs-verified: MCP stdio
Access docs-verified: Open-source, MIT licence (per registry metadata). Free local install from PyPI with python -m pip install --upgrade ras-commander-mcp. Requires Python 3.10+. No HEC-RAS executable needed. No auth listed in entry.

Useful for

Use cases llm-inferred: Check project length units and plan descriptions in existing HEC-RAS models during flood model QA · Pull selected plan metadata from HEC-RAS text files into a bounded review note for a model handover · Confirm ras-commander and ras-commander-mcp versions in a managed environment before starting modelling work
Audience llm-inferred: flood consultants · hydraulic modellers · hydrologists

Limitations

Constraints docs-verified: Read-only: does not modify projects, compute, read HDF/DSS, inspect geometry, extract gridded data, export files or download into projects · Project metadata and plan configuration need ras-commander 0.104.0; on 0.103.0 those tools return an availability error · Project tools are meant only for bounded informational subagents; other hosts need a scoped Python workflow · Only project roots listed in RAS_MCP_ALLOWED_ROOTS are readable; no roots are trusted by default · Claude Code plugin is forthcoming and its repository is not yet published · Version 0.4 replaces the 0.3 tool set and needs migration · Windows: other tools in a workflow may still fail on paths over 260 characters · Independent project, not affiliated with or endorsed by USACE HEC

flood hydrology

Some details auto-extracted · unverified

Globalhealthy

mcp

71/100Currency B

gaff-check ↗

MCP server exposing Ireland's public property datasets (planning, sold prices, flood, radon, zoning, BER context) as tools for AI agents

What's in it

Data offered docs-verified: address geocoding (OpenStreetMap Nominatim) · planning applications (National Planning Application Database, DHLGH) · sold prices (Property Price Register, PSRA) · radon risk areas (EPA Radon Risk Map) · generalised land-use zoning (Myplan, DHLGH) · groundwater and historic flood extents (GSI flood mapping) · BER typical-for-age context and SEAI grant reference data (CSO EBA02) · recorded-crime category counts for Garda sub-district and wider Garda area (CSO CJA11/CJA39/CJQ10/CJQ11)
Coverage docs-verified: Ireland
Formats docs-verified: MCP stdio
Access docs-verified: Free and open source; no API key needed; runs locally via npx (Node.js 20+). Code licence MIT; returned data is subject to underlying public dataset licences (mostly CC-BY 4.0 / PSI). Catalogue licence field says 'Other'.

Useful for

Use cases llm-inferred: Pre-acquisition screening of a site for flood, radon and zoning constraints · Planning-context scoping: nearby planning applications and zoning for a site appraisal · Sold-price context for a residential property in a feasibility or due-diligence note · Recorded-crime and BER area context as background in a property due-diligence brief
Audience llm-inferred: environmental consultants · flood risk consultants · planning consultants · property due-diligence analysts · GIS analysts

Limitations

Constraints docs-verified: Returns records only; not a valuation engine · Eircode-only lookup is not supported; an address or rough location is needed · River and coastal flood models are not included (licence-restricted; check floodinfo.ie) · Zoning is generalised land-use zoning (Myplan), not a detailed parcel-level designation · Recorded crime is area-level CSO counts, not incidents at an address; homicide and sexual offences are not published at sub-district level · Early release (v0.2.0); post-launch reliability and upstream-drift detection are active priorities · Requires Node.js 20+; running npx alone waits for an MCP client and appears to hang

flood planning

Some details auto-extracted · unverified

Irelandhealthy

dataset

70/100Currency A

Defra Data Services Platform LiDAR ↗

Environment Agency survey portal for LiDAR terrain and surface models and other survey data.

What's in it

Data offered llm-inferred: LiDAR terrain model · LiDAR surface model · other survey data
Coverage llm-inferred: England

Useful for

Use cases llm-inferred: Flood risk assessment: ground-level inputs for hydraulic model topography · Drainage strategy: surface levels for catchment and overland flow delineation · Civils design: existing ground model for earthworks and levels
Audience llm-inferred: flood consultants · drainage engineers · hydrologists · GIS analysts · civils engineers

Limitations

Constraints llm-inferred: No docs retrieved, so coverage, formats, resolution and access terms are unverified · Auth and licence not recorded in entry

terrain flood gis

Englandhealthy

mcp

70/100Currency A

mcp-fema-nfhl ↗

FEMA National Flood Hazard Layer (NFHL) MCP

flood environment

Some details auto-extracted · unverified

United Stateshealthy

mcp

70/100Currency A

mcp-flood ↗

MCP server for river discharge and flood forecasts via Open-Meteo

flood hydrology climate

Some details auto-extracted · unverified

Globalhealthy

mcp

70/100Currency A

mcp-uk-flood-monitoring ↗

UK Environment Agency real-time flood-monitoring MCP.

flood

Some details auto-extracted · unverified

United Kingdomhealthy

mcp

70/100Currency A

mcp-usda-soil ↗

USDA-NRCS Soil Data Access over SSURGO, the field-surveyed US soil map — the soil at any US coordinate, the named soil series inside a map unit with their taxonomy, drainage, slope and depth horizons, and SQL over the full survey schema.

hydrology geology gis flood

Some details auto-extracted · unverified

United Stateshealthy

mcp

68/100Currency B

local-dfw-mcp ↗

Open-source MCP server for DFW civic & property questions — Dallas 311, crime, flood zones, schools, districts, utilities — from authoritative city/county/state sources.

What's in it

Data offered docs-verified: local events (official city calendars; concerts, sports and theatre via Ticketmaster with a free key) · 311 service requests (City of Dallas only) · police incidents (Dallas, Fort Worth, Denton) · building permits (Fort Worth, McKinney, Arlington) · code-compliance violations (Fort Worth, McKinney, Arlington) · real-time traffic incidents (Fort Worth) · street and lane closures (Dallas, Arlington) · TxDOT annual traffic counts and construction projects (4 core counties) · FEMA flood zone with plain-English insurance interpretation · public schools and TEA A-F ratings (2022-23) · NWS active weather alerts · water/sewer provider (PUC CCN, Texas) · county, council district and ISD lookup (DFW) · county appraisal records (2025 certified roll)
Coverage docs-verified: Dallas-Fort Worth, Texas; coverage varies by tool (City of Dallas only for 311; Fort Worth default for permits and code cases; national for FEMA and NWS; Texas for schools and utilities) · TEA ratings 2022-23; appraisal 2025 certified roll; other sources not stated
Resolution docs-verified: Crime incidents are block-level and privacy-rounded upstream; resolution for other sources not stated
Update frequency docs-verified: Real-time for Fort Worth traffic incidents and NWS alerts; refresh cadence for other sources not stated
Formats llm-inferred: MCP stdio
Access docs-verified: Free and open; no API keys required for core tools. Optional free keys: DFW_TICKETMASTER_API_KEY (Ticketmaster, 5000 calls/day) and DFW_SODA_APP_TOKEN (higher Dallas open-data rate limit). Licence: Apache-2.0

Useful for

Use cases llm-inferred: Flood-zone screening for a DFW site address using dfw_fema_flood, with insurance interpretation · Development watch on a site: permits and code-compliance cases in Fort Worth, McKinney or Arlington · Utility provider identification (water/sewer PUC CCN) for a Texas site during due diligence · Council district, county and ISD lookup for a site to support planning and stakeholder mapping
Audience llm-inferred: flood risk consultants · planners · land and property due-diligence consultants · GIS analysts

Limitations

Constraints docs-verified: Covers Dallas-Fort Worth, Texas only; no UK data (entry regions field says Global, which the docs do not support) · Dallas permits and code cases are not wired (stale feeds; code-case publication stalled 2025-01-31) · Irving and other suburbs beyond Fort Worth, McKinney, Arlington, Denton and Dallas are not covered; composed property tool is pending · Not for tenant, employment, credit, insurance or other FCRA-regulated screening · Appraisal values are the 2025 certified appraised value, not a tax bill; only 4 core counties verified · Events from Ticketmaster require a free API key; some suburbs publish no usable calendar · Requires Node 20 or later · Upstream free text (311, events) is untrusted and treated as data

flood

Some details auto-extracted · unverified

Globalhealthy

mcp

68/100Currency A

rider-mcp-enforcer ↗

Claude Code plugin: force Rider's MCP symbol/reference/file index over Bash grep, with token-flood summarization. For large Unreal C++/.NET codebases.

What's in it

Data offered docs-verified: symbol search (search_symbol) · file search (search_file) · text and regex search (search_text, search_regex) · file lookup by name or glob (find_files_by_name_keyword, find_files_by_glob) · symbol info (get_symbol_info) · rename refactoring (rename_refactoring) · read file (read_file)
Formats docs-verified: MCP
Access docs-verified: Free, open source (MIT licence); requires JetBrains Rider 2025.2+ running with its MCP server connected

Useful for

Use cases llm-inferred: Locate a C# class or symbol definition in a large .NET codebase without grep flooding the context · Find string references to a symbol before a change using search_text (not semantic find-usages) · Rename a C# symbol across a project with rename_refactoring instead of sed
Audience llm-inferred: software developers using JetBrains Rider on C#/.NET codebases · game developers working on large Unreal C++ projects

Limitations

Constraints llm-inferred: Not a water, civils or environmental data source; it is a developer code-navigation tool · Requires Rider running with its MCP server connected · No semantic find-usages/find-references tool; reference search is indexed string match · search_symbol on Unreal C++ can return filename or path matches instead of the exact class · Results are capped at the top N (RIDER_MAX_RESULTS), so output is not exhaustive · Rider's index excludes Saved/ (logs, build output), so log paths return empty results · rider_regen_project auto-detect is Windows-only, and Rider must reload the solution after regen

flood

Some details auto-extracted · unverified

Globalhealthy

mcp

67/100Currency A

hydrata-mcp-server ↗

MCP server for Hydrata — AI agent interface for ANUGA flood simulation

flood

Some details auto-extracted · unverified

Globalhealthy

mcp

67/100Currency A

mcp-json-lens ↗

MCP server and CLI that inspect and query large JSON and JSONL files with jq-like expressions - compact results, no context flooding.

flood

Some details auto-extracted · unverified

Globalhealthy

mcp

66/100Currency B

Agentic-MIKE-Plus ↗

Skills + MCP server for Claude Code, Codex, Hermes, or OpenClaw - headless MIKE+ automated modelling and analysis (inspect, edit, run, read, plot), no GUI.

What's in it

Data offered docs-verified: MIKE+ model overview (simulations, scenarios, model type, element counts) · Scenario activation · Parameter read/write (e.g. pipe diameter) · Headless simulation run with QA status (completed/errors/warnings) · SWMM/EPANET .inp import into MIKE+ · Flooding node ranking (peak water level vs ground level) · Baseline vs scenario comparison (delta peak/volume/timing, RMSE, NSE) · Rainfall CSV to MIKE+ .dfs0 · Hydrograph, profile and network plots · Provenance manifest with sha256
Formats docs-verified: MIKE+ .dfs0 · MIKE+ .res1d · EPA SWMM .inp · EPANET .inp · CSV (rainfall input)
Access docs-verified: MIT licence (software). Running, editing and importing require a valid DHI MIKE+ license; reading, plotting and comparing need no MIKE+ license.

Useful for

Use cases llm-inferred: Baseline vs upgraded drainage network scenario runs with mike_run and mike_results_compare · Flood-node screening by ranking peak water level against ground level from a MIKE+ run · Converting design-storm rainfall CSVs to MIKE+ .dfs0 inputs · Importing SWMM/EPANET models into MIKE+ and writing provenance manifests for QA
Audience llm-inferred: drainage engineers · flood consultants · hydrologists

Limitations

Constraints docs-verified: Experimental / pre-release status · Run, edit and import need a licensed MIKE+ 2026 install and Windows with Python 3.11 x64 · Verified end to end only on the MIKE+ 2026 Sirius_RTC example · mikeplus requires Python 3.9-3.11, so Python 3.12+ is not supported

hydrology flood

Some details auto-extracted · unverified

Globalhealthy

mcp

66/100Currency B

gagelink ↗

MCP server giving AI agents hydrology data from USGS, NOAA, France's Hub'Eau, the UK Environment Agency and SWOT: river level, streamflow, flood forecasts, basins. Values carry their unit, datum and record quality, so a stage is never silently compared against an elevation.

flood hydrology climate gis terrain

Some details auto-extracted · unverified

United Kingdom · United States · Francehealthy

mcp

66/100Currency B