swmm-runnerlisted

Run EPA SWMM (swmm5) simulations reproducibly and extract key metrics from the report file. Use when an agent needs to (1) run a .inp via swmm5 CLI, (2) generate a run directory with rpt/out + manifest, (3) extract peak flow/time for a node/outfall, (4) parse SWMM continuity (Runoff Quantity / Flow Routing) errors from .rpt, or (5) compare two .rpt files (e.g. GUI vs CLI) for equivalence.
Zhonghao1995/agentic-swmm-workflow · ★ 8 · Data & Documents · score 71

Install: claude install-skill Zhonghao1995/agentic-swmm-workflow

# SWMM Runner (CLI-first) ## What this skill provides - Deterministic execution wrapper around the `swmm5` binary. - A standard **run directory** layout: `inp/`, `rpt/`, `out/`, `stdout`, `stderr`, `manifest.json`. - Metric extraction read directly from SWMM's own `.rpt`: - peak flow + time-of-peak for any junction or outfall; - Runoff Quantity and Flow Routing continuity tables + continuity error %; - cross-run comparison (e.g. GUI vs CLI) on continuity error. ## When to use this skill Use after a SWMM model is fully assembled (typically by `swmm-builder.build_inp`) and you need to actually execute it and read back the metrics. Also use to compare two .rpt files for regression / GUI parity. Do **not** use this skill to assemble the .inp itself (that's `swmm-builder`) or to plot the results (that's `swmm-plot`). ## MCP tools `mcp/swmm-runner/server.js` exposes four tools. 1. **`swmm_run`** — run `swmm5` against an .inp and write rpt + out + stdout + stderr + manifest.json into a run directory. - Args: `inp` (path), `runDir` (path), `node` (optional), `rptName` (optional), `outName` (optional). - When `node` is omitted, the server auto-detects the first entry from the .inp `[OUTFALLS]` section so the manifest's peak metric targets a real outfall name (no more silent "O1" default). - Output: a manifest with inp_sha256, swmm5 version, file paths, `metrics.peak`, `metrics.continuity`. 2. **`swmm_peak`** — parse peak flow and time-of-peak for a specific nod