# agentic-swmm-workflow - Doramagic AI Context Pack > Positioning: a pre-install experience and judgment asset. It helps the host AI get off to a good start, but it does not mean the project has already been installed, run, or validated. ## Sufficiency Principle - **Sufficiency over compression**: The AI Context Pack should be sufficient for the host AI to understand the project's value, capability boundaries, entrypoints, risks, and evidence sources before starting work; it may be layered, but it does not aim for the shortest possible summary. - **Compression policy**: Compress only noise and duplication, never context that affects judgment or the quality of the work. ## How the Host AI Should Use This You are reading the AI Context Pack that Doramagic compiled for agentic-swmm-workflow. Treat it as pre-work context: help the user understand who it fits, what it can do, how to start, what must be verified after install, and where the risks are. Do not claim that you have already installed, run, or executed the target project. ## Claim Consumption Rules - **Fact source**: Repo Evidence + Claim/Evidence Graph; the Human Wiki only supplies salience, terminology, and narrative structure. - **Minimum status for a fact**: `supported` - `supported`: May be used as a project fact, but the answer must cite the claim_id and evidence path. - `weak`: Usable only as a low-confidence lead; the user must be asked to keep verifying. - `inferred`: Usable only for risk notes or open questions; must not be packaged as a project fact. - `unverified`: Must not be used as fact; state clearly that evidence is insufficient. - `contradicted`: Must show the conflicting sources and must not force a single version on the user's behalf. ## Who It Fits Best - **AI researchers or builders of research-oriented Agents**: The README clearly centers on research, experiment, or paper workflows. Evidence: `README.md` Claim: `clm_0003` supported 0.86 - **Developers already using host AIs such as Claude/Codex/Cursor/Gemini**: The README or plugin config mentions multiple host AIs. Evidence: `README.md` Claim: `clm_0004` supported 0.86 - **Users who want to bring professional workflows into a host AI**: The repo contains Skill documents. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. Claim: `clm_0005` supported 0.86 ## What It Can Do - **AI Skill / Agent Instruction Asset Library** (Previewable before install): The project contains Skill or Agent instruction files that a host AI can read, useful for bringing professional workflows into hosts like Claude, Codex, or Cursor. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Command-Line Startup or Install Flow** (Verify after install): The project documentation contains runnable commands; real use requires running them in a local or host environment. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` Claim: `clm_0002` supported 0.86 ## How to Start - `curl -fsSL https://aiswmm.com/install.sh | bash` Evidence: `README.md` Claim: `clm_0006` supported 0.86 - `npx skills add Zhonghao1995/agentic-swmm-workflow` Evidence: `README.md` Claim: `clm_0007` supported 0.86 - `pip install "aiswmm[anywhere]"` Evidence: `docs/swmm-anywhere-quickstart.md` Claim: `clm_0008` supported 0.86 ## Continue-or-Stop Decision Card - **Current recommendation**: Sandbox-trial permissions first - **Why**: The project has signals of install commands, host configuration, or local writes; do not go straight into your primary environment—trial it in isolation first. ### 30-Second Read - **What to do now**: Sandbox-trial permissions first - **Minimum safe next step**: Run Prompt Preview first; if you still want to install, trial only in an isolated environment - **Do not trust yet**: Tool permission boundaries cannot be trusted before install. - **Continuing will touch**: Command execution, Host AI configuration, Local environment or project files ### What You Can Trust Now - **Target-audience signal: AI researchers or builders of research-oriented Agents** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `README.md` Claim: `clm_0003` supported 0.86 - **Target-audience signal: Developers already using host AIs such as Claude/Codex/Cursor/Gemini** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `README.md` Claim: `clm_0004` supported 0.86 - **Target-audience signal: Users who want to bring professional workflows into a host AI** (supported): Backed by a supported claim or project evidence, but that still is not the same as real install results. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. Claim: `clm_0005` supported 0.86 - **Capability exists: AI Skill / Agent Instruction Asset Library** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Capability exists: Command-Line Startup or Install Flow** (supported): You can trust that the project contains signals of this capability; whether it fits your specific task still needs trial or after-install verification. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` Claim: `clm_0002` supported 0.86 - **There are Quick Start / install-command signals** (supported): You can trust that the docs mention a startup or install entrypoint; do not run it directly in your primary environment because of that. Evidence: `README.md` Claim: `clm_0006` supported 0.86 ### What You Cannot Trust Yet - **Tool permission boundaries cannot be trusted before install.** (unverified): MCP/tool projects usually touch files, the network, the browser, or external APIs, so permissions and logs must be checked for real. - **Real output quality cannot be trusted before install.** (unverified): Prompt Preview can only show how it guides you; it cannot prove result quality in the real project. - **Host AI version compatibility cannot be trusted before install.** (unverified): Host loading rules and version differences across Claude, Cursor, Codex, Gemini, and others must be verified in a real environment. - **That it will not pollute your existing host AI's behavior cannot be trusted directly.** (inferred): Skill, plugin, and AGENTS/CLAUDE/GEMINI instructions may change the host AI's default behavior. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. - **Safe rollback cannot be assumed by default.** (unverified): Unless the project clearly provides uninstall and recovery instructions, verify in an isolated environment first. - **After a real install, is it compatible with the user's current host AI version?** (unverified): Compatibility can only be verified in the actual host environment. - **Does the project's output quality meet the user's specific task?** (unverified): The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. - **Do the install commands require network access, permissions, or global writes?** (unverified): This affects install risk in both enterprise and personal environments. Evidence: `README.md` ### What Continuing Will Touch - **Command execution**: Package managers, network downloads, the local plugin directory, project config, or the user's home directory. Why: Running the very first command can already change your environment; decide whether it is worth running first. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` - **Host AI configuration**: The plugin, Skill, or rule-loading config of hosts like Claude/Codex/Cursor/Gemini/OpenCode. Why: Host configuration changes how the AI works afterward and may conflict with the user's existing rules. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. - **Local environment or project files**: Install results, plugin caches, project config, or local dependency directories. Why: The write scope and rollback path cannot be proven before install and need isolated verification. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` - **Host AI context**: The AI Context Pack, Prompt Preview, Skill routing, risk rules, and project facts. Why: Importing context affects the host AI's later judgment, so avoid packaging unverified items as facts. ### Minimum Safe Next Steps - **Run Prompt Preview first**: Use a pre-install interactive trial to judge whether the way of working fits; it needs no authorization or environment change. (applies when: Applies to any project, especially when output quality is unknown.) - **Trial-install only in an isolated directory or a test account**: Avoid letting install commands pollute your primary host AI, real projects, or home directory. (applies when: When there are signals of command execution, plugin config, or local writes.) - **Back up your host AI configuration first**: Skill, plugin, and rule files may change the default behavior of Claude/Cursor/Codex. (applies when: When there is a plugin manifest, a Skill, or a host rule entrypoint.) - **After install, verify just one minimal task**: Verify loading, compatibility, output quality, and rollback first, then decide whether to use it deeply. (applies when: When moving from a trial into a real workflow.) ### Exit Plan - **Preserve the pre-install state**: Record the original host config and project state so you can later judge whether it is recoverable. - **Be ready to remove the host plugin / Skill / rule entrypoint**: If behavior is off after the trial install, you can restore the host AI to its pre-trial state. - **Record the install commands and written paths**: Without clear uninstall instructions, you at least need to know which directories or configs to clean up manually. - **If there is no rollback path, do not enter your primary environment**: No rollback is a blocker before continuing; do not proceed on trust or luck. ## What Can Only Be Previewed - Explain who the project fits and what it can do - Demonstrate a typical conversation flow based on project docs - Help the user decide whether it is worth installing or researching further ## What Must Be Verified After Install - Actually installing the Skill, plugin, or CLI - Running scripts, modifying local files, or accessing external services - Verifying real output quality, performance, and compatibility ## Boundary & Risk Decision Card - **Mistaking the pre-install preview for a real run**: The user may overestimate how much configuration, permission, and compatibility verification the project has already done. Mitigation: Clearly separate prompt_preview_can_do from runtime_required. Claim: `clm_0009` inferred 0.45 - **Command execution will modify the local environment**: Install commands may write to the user's home directory, the host plugin directory, or project configuration. Mitigation: Run in an isolated environment or a test account first. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` Claim: `clm_0010` supported 0.86 - **Source document conflict: skill_count**: The project documentation states inconsistent counts; the AI Context Pack must warn the user not to treat any single number as a verified fact. Mitigation: Flag it as unverified in both the Human Manual and the AI Context Pack rather than forcing a single number. Evidence: `docs/agent-nl-swmm-evidence-20260514.md`, `docs/agent-nl-swmm-gpt55-evidence-20260515.md`, `CHANGELOG.md` Claim: `clm_0011` supported 0.86, `clm_0012` contradicted 0.20 - **Source file conflict skill_count**: multiple values `14, 18` found; verify before real use. - **To confirm**: After a real install, is it compatible with the user's current host AI version?. Why: Compatibility can only be verified in the actual host environment. - **To confirm**: Does the project's output quality meet the user's specific task?. Why: The pre-install preview can only show flow and boundaries; it cannot replace real evaluation. - **To confirm**: Do the install commands require network access, permissions, or global writes?. Why: This affects install risk in both enterprise and personal environments. ## Pre-Work Working Context ### Loading Order - First read how_to_use.host_ai_instruction to establish the boundaries of this pre-install judgment asset. - Read claim_graph_summary to confirm facts come from the Claim/Evidence Graph, not the Human Wiki narrative. - Then read intended_users, capabilities, and quick_start_candidates to judge whether the user is a match. - When you need to carry out a concrete task, check role_skill_index first, then evidence_index. - For real install, file modification, network access, performance, or compatibility questions, turn to risk_card and boundaries.runtime_required. ### Task Routes - **AI Skill / Agent Instruction Asset Library**: Use role_skill_index / evidence_index to help the user pick a usable role, Skill, or workflow first. Boundary: Can be experienced via a pre-install Prompt. Evidence: `skills/skill-author/SKILL.md`, `skills/swmm-anywhere/SKILL.md`, `skills/swmm-builder/SKILL.md`, `skills/swmm-calibration/SKILL.md` et al. Claim: `clm_0001` supported 0.86 - **Command-Line Startup or Install Flow**: State that this is an after-install capability first, then give a pre-install checklist. Boundary: Must be verified after a real install or run. Evidence: `README.md`, `docs/swmm-anywhere-quickstart.md` Claim: `clm_0002` supported 0.86 ### Context Scale - Total files: 510 - Important-file coverage: 40/510 - Evidence index entries: 80 - Role / Skill entries: 18 ### Handling Insufficient Evidence - **missing_evidence**: State that evidence is insufficient and ask the user for the target file, a README section, or after-install verification records; do not fill in facts. - **out_of_scope_request**: State that the task is beyond the current AI Context Pack's evidence scope and suggest the user check the Human Manual or verify after a real install. - **runtime_request**: Provide a pre-install checklist and command sources, but do not run commands for the user or claim they have been run. - **source_conflict**: Show the conflicting sources side by side, mark them as unverified, and do not force a single version. ## Prompt Recipes ### Fit assessment - Goal: Judge whether this project fits the user's current task. - Expected output: A fit conclusion, key reasons, evidence citations, what can be previewed before install, what must be verified after install, and a next-step recommendation. ```text Based on the AI Context Pack for agentic-swmm-workflow, ask me 3 necessary questions first, then judge whether it fits my task. The answer must cover: who it fits, what it can do, what it cannot do, whether it is worth installing, and where the evidence comes from. Every project fact must cite evidence_refs, source_paths, or a claim_id. ``` ### Pre-install experience - Goal: Let the user feel the core workflow before installing, while avoiding packaging the preview as real capability or a marketing promise. - Expected output: An experience script with boundary labels, an after-install verification checklist, and a cautious recommendation; with no real-run promises or strong marketing language. ```text Treat agentic-swmm-workflow as a pre-install experience asset, not an already-installed tool or a real runtime environment. Output exactly four parts: 1. Ask me 3 necessary questions first. 2. Give an "experience script": use the three labels [Previewable before install], [Must verify after install], and [Insufficient evidence] to show how it might guide the workflow. 3. Give an after-install verification checklist: list which capabilities can only be confirmed after a real install, real host loading, and a real project run. 4. Give a cautious recommendation: only "worth researching/trialing further", "add information before deciding", or "not recommended to continue"; do not endorse the project. Hard boundaries: - Do not claim you have installed, run, executed tests, modified files, or produced real results. - Do not write promise-like phrasing such as "auto-adapts", "guarantees passing", "perfect fit", or "strongly recommend installing". - If you describe how it works after install, you must use a conditional such as "if installed successfully and the host loads the Skill correctly, it might...". - The experience script may only be written as "example lines / hypothetical flow": use "might ask / might suggest / might show", not "has written, has generated, has passed, is running, is generating". - Prompt Preview does not hand out install commands; if the user is ready to trial, only prompt them to read Quick Start and the Risk Card first and to verify in an isolated environment. - Every project fact must come from a supported claim, evidence_refs, or source_paths; inferred/unverified items can only be risks or open questions. ``` ### Role / Skill selection - Goal: Pick the best-matching asset from the project's roles or Skills. - Expected output: A list of candidate roles or Skills, each with an applicable scenario, evidence paths, risk boundary, and whether after-install verification is needed. ```text Read role_skill_index and recommend 3-5 of the most relevant roles or Skills for my target task. For each recommendation, state the applicable scenario, likely output, risk boundary, and evidence_refs. ``` ### Risk pre-check - Goal: Identify environment, permission, rule-conflict, and quality risks before installing or adopting. - Expected output: A checklist of environment, permission, dependency, license, host-conflict, quality risk, and unknown items. ```text Based on risk_card, boundaries, and quick_start_candidates, give me a pre-install risk pre-check list. Do not run commands for me; only explain what I should check, why, and what impact a failure would have. ``` ### Host AI kickoff instruction - Goal: Turn the project context into a host AI instruction for the start of a conversation. - Expected output: A pre-work instruction with clear boundaries and clear evidence citations, suitable to copy to a host AI. ```text Based on the AI Context Pack for agentic-swmm-workflow, generate a pre-work instruction I can paste to my host AI. This instruction must obey not_runtime=true and must not claim the project has been installed, run, or produced real results. ``` ## Role / Skill Index - Indexed 18 role / Skill / project-doc entries. - **skill-author** (skill): Draft a well-formed new skill a SKILL.md scaffold, optionally with scripts/references from a described recurring need, for human review and approval. Use whenever a repeated workflow gap has no existing skill covering it, when someone wants to propose or create a new skill or capability, or when an agentic system detects a recurring problem that warrants a brand-new skill rather than editing an existing one. Domain-… Activation hint: When the user's task is highly relevant to the workflow described by “skill-author”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/skill-author/SKILL.md` - **swmm-anywhere** (skill): Synthesize a plausible SWMM drainage network from public data OSM streets + DEM when NO real pipe-network data exists — input is just a bbox. Use ONLY when the user has no pipe shapefile/CAD/GIS data, or to establish a baseline before real data arrives; if real pipe data exists, route to swmm-network or swmm-gis instead. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-anywhere”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-anywhere/SKILL.md` - **swmm-builder** (skill): Assemble a runnable SWMM INP deterministically from subcatchment geometry/attributes, merged parameter JSON, network JSON, and climate references. Use when creating auditable INP + manifest artifacts for downstream swmm-runner/calibration. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-builder”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-builder/SKILL.md` - **swmm-calibration** (skill): Calibration and validation scaffold for EPA SWMM. Use when an agent needs to 1 compare simulated vs observed flow, 2 evaluate candidate parameter sets, 3 rank explicit candidates by an objective, 4 run a bounded random / LHS / adaptive search for the best-fitting parameters, 5 run a publication-grade SCE-UA calibration with KGE as the primary objective and r, alpha, beta decomposition reported, or 6 run a DREAM-ZS B… Activation hint: When the user's task is highly relevant to the workflow described by “swmm-calibration”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-calibration/SKILL.md` - **swmm-climate** (skill): Deterministic rainfall/climate formatting for SWMM. Use when converting timestamped rainfall CSV files into SWMM-ready TIMESERIES lines and RAINGAGES helper snippets for swmm-builder. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-climate”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-climate/SKILL.md` - **swmm-design-review** (skill): Activation hint: When the user's task is highly relevant to the workflow described by “swmm-design-review”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-design-review/SKILL.md` - **swmm-end-to-end** (skill): Top-level orchestration skill for agentic SWMM modelling. Use when an agent needs one entrypoint that decides which module tools to run, in what order, and when to stop, for example to build, run, QA, and optionally calibrate a SWMM case from prepared or partially prepared inputs. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-end-to-end”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-end-to-end/SKILL.md` - **swmm-experiment-audit** (skill): Consolidate Agentic SWMM run artifacts into auditable provenance, comparison records, and local Obsidian audit notes. Use after any SWMM build/run/QA attempt, successful or failed, when an agent or CLI workflow needs a traceable record of inputs, commands, artifacts, metrics, QA checks, run-to-run differences, and first-user-friendly Obsidian visualization. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-experiment-audit”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-experiment-audit/SKILL.md` - **swmm-gis** (skill): GIS/DEM preprocessing for SWMM experiments using the user's own QGIS/GRASS layers. Use when the user asks to 1 delineate subcatchments through QGIS/GRASS standard or entropy-guided , 2 preprocess QGIS-derived subcatchment polygons into builder-ready CSV, 3 identify high-entropy hotspot subcatchments, or 4 expose QGIS/GRASS-backed preprocessing as MCP tools for reproducible workflows. For bbox-only inputs WITHOUT rea… Activation hint: When the user's task is highly relevant to the workflow described by “swmm-gis”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-gis/SKILL.md` - **swmm-modeling-memory** (skill): Read historical Agentic SWMM experiment audit artifacts and summarize repeated assumptions, QA issues, failures, missing evidence, run-to-run differences, lessons learned, and controlled skill update proposals. Use downstream of swmm-experiment-audit when multiple audited runs exist or when a user asks for modeling memory, failure-pattern extraction, lessons learned, or human-reviewed skill refinement proposals. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-modeling-memory”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-modeling-memory/SKILL.md` - **swmm-network** (skill): Build, validate, and route SWMM pipe-network models for urban drainage from raw municipal shapefiles or structured GIS/CAD exports. Use when handling junctions, conduits, outfalls, xsections, network field-mapping configs, or wiring subcatchments to upstream nodes. Requires real pipe data as SHP / GeoJSON / CSV — native CAD DXF/DWG is not parsed and must first be exported to one of these. For data-scarce areas where… Activation hint: When the user's task is highly relevant to the workflow described by “swmm-network”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-network/SKILL.md` - **swmm-params** (skill): Deterministic mapping from land use and soil texture to SWMM runoff/subarea and Green-Ampt infiltration parameters. Use when generating first-pass subcatchment parameter tables for swmm-builder. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-params”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-params/SKILL.md` - **swmm-plot** (skill): Publication-grade plotting for SWMM rainfall–runoff time-series figures. Use when an agent needs to produce a paired rainfall top, inverted + node flow bottom figure from a SWMM .inp + .out, with strict style rules SI units, Arial, ticks inward, no title , optionally cropped to an event window or focus day. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-plot”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-plot/SKILL.md` - **swmm-rag-memory** (skill): Retrieve relevant Agentic SWMM modeling memory from audited runs, modeling-memory summaries, and Obsidian-compatible notes at query time. Use when a user asks for RAG, similar past runs, evidence-linked memory retrieval, historical QA/failure patterns, or memory-grounded answers. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-rag-memory”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-rag-memory/SKILL.md` - **swmm-report** (skill): Activation hint: When the user's task is highly relevant to the workflow described by “swmm-report”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-report/SKILL.md` - **swmm-runner** (skill): 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. Activation hint: When the user's task is highly relevant to the workflow described by “swmm-runner”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-runner/SKILL.md` - **swmm-uncertainty** (skill): Parameter and forcing uncertainty propagation and sensitivity analysis for EPA SWMM. Use when an agent needs to 1 propagate parameter uncertainty through SWMM fuzzy alpha-cut or Monte Carlo , 2 quantify hydrograph envelopes or output entropy without treating the run as calibration, 3 screen which parameters matter using OAT / Morris elementary-effects / Sobol' indices, 4 generate a rainfall ensemble observed-series… Activation hint: When the user's task is highly relevant to the workflow described by “swmm-uncertainty”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-uncertainty/SKILL.md` - **swmm-water-quality** (skill): Activation hint: When the user's task is highly relevant to the workflow described by “swmm-water-quality”, use it for a pre-install experience first, then decide whether to install. Evidence: `skills/swmm-water-quality/SKILL.md` ## Evidence Index - Indexed 80 evidence entries. - **Saanich E2E with B7 — pipe network actually used 2026-05-13** (documentation): Saanich E2E with B7 — pipe network actually used 2026-05-13 Evidence: `docs/framework-validation/saanich-b7-network-routed-20260513/README.md` - **Saanich E2E with B8 — water finally reaches OUT1 2026-05-13** (documentation): Saanich E2E with B8 — water finally reaches OUT1 2026-05-13 Evidence: `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/README.md` - **Cold-start agent validation — Saanich Cecelia Ravine 2026-05-13** (documentation): Cold-start agent validation — Saanich Cecelia Ravine 2026-05-13 Evidence: `docs/framework-validation/saanich-cold-start-cecelia-20260513/README.md` - **Saanich end-to-end with new MCP tools — 2026-05-13 lock-in** (documentation): Saanich end-to-end with new MCP tools — 2026-05-13 lock-in Evidence: `docs/framework-validation/saanich-e2e-new-tools-20260513/README.md` - **Saanich MCP-first framework smoke — 2026-05-13 lock-in** (documentation): Saanich MCP-first framework smoke — 2026-05-13 lock-in Evidence: `docs/framework-validation/saanich-smoke-20260513/README.md` - **Agentic SWMM Workflow** (documentation): Pre-1.0 · stable v0.7.4 · pip install aiswmm==0.7.4 · CHANGELOG CHANGELOG.md 🚧 In active development: SWMMCanada https://github.com/Zhonghao1995/SWMMCanada , a SWMM INP generator built from Canadian open data: draw anywhere in Canada and run, with real municipal storm networks for 7 cities or synthesized everywhere else. Evidence: `README.md` - **Agentic SWMM Integrations** (documentation): This directory contains runtime integration guidance for agent systems that use Agentic SWMM outside the direct CLI path. Evidence: `integrations/README.md` - **Local run outputs** (documentation): runs/ is for local generated outputs. Do not commit run artifacts from this folder. Evidence: `runs/README.md` - **Agentic SWMM Website Installers** (documentation): These files are meant to be hosted at the root of the public website: Evidence: `web/README.md` - **Agentic SWMM Public Agent Memory Layer** (documentation): Agentic SWMM Public Agent Memory Layer Evidence: `agent/memory/README.md` - **Calibration example MVP** (documentation): This folder contains a minimal example configuration for the public calibration scaffold. Evidence: `examples/calibration/README.md` - **Tecnopolo Prepared-Input Benchmark** (documentation): This example is a compact external benchmark for the Agentic SWMM prepared-input workflow. It is derived from the public Zenodo Tecnopolo SWMM dataset and trimmed to January 1994 so it can be committed and rerun quickly. Evidence: `examples/tecnopolo/README.md` - **TUFLOW SWMM Module 03 Raw GeoPackage Benchmark** (documentation): TUFLOW SWMM Module 03 Raw GeoPackage Benchmark Evidence: `examples/tuflow-swmm-module03/README.md` - **MCP Runtime Integration** (documentation): This folder makes the Agentic SWMM module MCP servers easier to attach to Codex, Hermes, OpenClaw, or any other stdio MCP client. Evidence: `integrations/mcp/README.md` - **Skill Installation** (documentation): Agentic SWMM ships its orchestration knowledge as repository skills under skills/ . Evidence: `integrations/skills/README.md` - **SWMM Modeling Memory Example** (documentation): Run modeling-memory summarization after one or more run directories have been audited by swmm-experiment-audit . Evidence: `skills/swmm-modeling-memory/examples/README.md` - **City Dual-System Structured Network Benchmark** (documentation): City Dual-System Structured Network Benchmark Evidence: `skills/swmm-network/examples/city-dual-system/README.md` - **mapping.json templates for import city network** (documentation): mapping.json templates for import city network Evidence: `skills/swmm-network/templates/README.md` - **Skill author** (skill_instruction): Turn a description of a recurring need into a draft skill that a human can review and approve. This skill writes proposals; it never installs, activates, or edits skills on its own — a freshly drafted skill is a proposal, not a verified capability. Evidence: `skills/skill-author/SKILL.md` - **swmm-anywhere** (skill_instruction): Synthesize a plausible SWMM drainage network from public data OSM streets + DEM when no real pipe-network data exists. Evidence: `skills/swmm-anywhere/SKILL.md` - **SWMM Builder INP assembly layer** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-builder/SKILL.md` - **SWMM Calibration / Validation** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-calibration/SKILL.md` - **SWMM Climate MVP rainfall layer** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-climate/SKILL.md` - **swmm-design-review — Design Review / Code-Compliance Checker** (skill_instruction): swmm-design-review — Design Review / Code-Compliance Checker Evidence: `skills/swmm-design-review/SKILL.md` - **SWMM End-to-End Orchestration** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-end-to-end/SKILL.md` - **SWMM Experiment Audit** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-experiment-audit/SKILL.md` - **SWMM GIS / Preprocess** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-gis/SKILL.md` - **SWMM Modeling Memory** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-modeling-memory/SKILL.md` - **SWMM Network pipe-system layer** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-network/SKILL.md` - **SWMM Params MVP mapping layer** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-params/SKILL.md` - **SWMM Plot publication spec** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-plot/SKILL.md` - **SWMM RAG Memory** (skill_instruction): - Query-time retrieval over Agentic SWMM audited run memory. - A lightweight keyword/tag retriever that works without embeddings or a vector database. - A local hybrid retriever that combines keyword matches, deterministic SWMM tags, metadata weighting, and hashed token/character n-gram embeddings. - RAG context packs that can be passed to Codex, OpenClaw, Hermes, or another LLM. - Source citations for each retrieved memory item, including run id, project key, source file, failure patterns, diagnostics, and matched terms. - Retrieval-grounded failure advice.{json,md} for failed or warning runs, without modifying model files. - Explicit resolution memory.json for human-reviewed and benchmark… Evidence: `skills/swmm-rag-memory/SKILL.md` - **SWMM Report Export Skill** (skill_instruction): Assemble a reproducible, client-deliverable Word .docx report from the artifacts produced by swmm-experiment-audit and swmm-plot . The script reads only existing files; it never re-runs SWMM or modifies the run directory. Evidence: `skills/swmm-report/SKILL.md` - **SWMM Runner CLI-first** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-runner/SKILL.md` - **SWMM Uncertainty** (skill_instruction): Part of Agentic SWMM https://github.com/Zhonghao1995/agentic-swmm-workflow — install the project first for the executable toolchain aiswmm CLI, SWMM solver, MCP servers . Evidence: `skills/swmm-uncertainty/SKILL.md` - **SWMM Water Quality Skill** (skill_instruction): Complete SWMM engine coverage for pollutant buildup/washoff simulation and load reporting. This skill provides: Evidence: `skills/swmm-water-quality/SKILL.md` - **API Key Configuration** (documentation): Agentic SWMM plans the interactive aiswmm runtime with one of two API-key providers : openai the default; OPENAI API KEY and anthropic opt-in; ANTHROPIC API KEY . Configure the relevant key during installation or in your shell environment before starting the runtime. For how to switch providers and which models each uses, see llm providers.md llm providers.md . Evidence: `docs/api-key-configuration.md` - **Byte-identical SWMM reproducibility across environments Tecnopolo** (documentation): Byte-identical SWMM reproducibility across environments Tecnopolo Evidence: `docs/byte-identical-reproducibility.md` - **Codex Runtime Path** (documentation): This document explains how to use Codex as the primary local development runtime for Agentic SWMM. Evidence: `docs/codex-runtime.md` - **Experiment Audit Framework** (documentation): This document defines the audit layer for Agentic SWMM. Evidence: `docs/experiment-audit-framework.md` - **Installation and CLI Guide** (documentation): This page keeps detailed setup notes out of the README. For most users, Docker is the recommended path because it keeps the SWMM solver and Python environment reproducible. Use local installation when you need to edit skills, run MCP servers, or develop the Python CLI. Evidence: `docs/installation.md` - **LLM Providers** (documentation): The aiswmm interactive runtime drives its planner with a large language model. Two API-key backends are supported, both using standard function-calling. The deterministic rule planner needs no provider at all; everything below applies only to the LLM planner. Evidence: `docs/llm_providers.md` - **OpenClaw Execution Path** (documentation): This document defines the intended top-level execution path for swmm-end-to-end when used by Codex, OpenClaw, Hermes, or another MCP-centered external agent runtime. Evidence: `docs/openclaw-execution-path.md` - **Runtime Install Options** (documentation): There look like four ways in, but they are three strategies — the two one-line installers are the same installer, shipped as a macOS/Linux build install.sh and a Windows build install.ps1 . Pick one: Evidence: `docs/runtime-install-options.md` - **Quickstart: rapid SWMM modelling in a data-scarce region via SWMManywhere + aiswmm audit** (documentation): Quickstart: rapid SWMM modelling in a data-scarce region via SWMManywhere + aiswmm audit Evidence: `docs/swmm-anywhere-quickstart.md` - **Package** (package_manifest): { "name": "swmm-builder-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-builder/package.json` - **Package** (package_manifest): { "name": "swmm-calibration-mcp", "version": "0.1.0", "private": true, "type": "module", "dependencies": { "@modelcontextprotocol/sdk": "^1.17.5", "zod": "^3.24.1" }, "scripts": { "start": "node server.js" } } Evidence: `mcp/swmm-calibration/package.json` - **Package** (package_manifest): { "name": "swmm-climate-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-climate/package.json` - **Package** (package_manifest): { "name": "swmm-experiment-audit-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-experiment-audit/package.json` - **Package** (package_manifest): { "name": "swmm-gis-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-gis/package.json` - **Package** (package_manifest): { "name": "swmm-modeling-memory-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-modeling-memory/package.json` - **Package** (package_manifest): { "name": "swmm-network-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-network/package.json` - **Package** (package_manifest): { "name": "swmm-params-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-params/package.json` - **Package** (package_manifest): { "name": "swmm-plot-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-plot/package.json` - **Package** (package_manifest): { "name": "swmm-runner-mcp", "version": "0.1.0", "type": "module", "main": "server.js", "scripts": { "start": "node server.js" }, "dependencies": { "@modelcontextprotocol/sdk": "^1.27.1", "zod": "^4.3.6" } } Evidence: `mcp/swmm-runner/package.json` - **Package** (package_manifest): { "name": "swmm-uncertainty-mcp", "version": "0.1.0", "private": true, "type": "module", "dependencies": { "@modelcontextprotocol/sdk": "^1.17.5", "zod": "^3.24.1" }, "scripts": { "start": "node server.js" } } Evidence: `mcp/swmm-uncertainty/package.json` - **License** (source_file): Copyright c 2026 Zhonghao Zhang & Caterina Valeo Evidence: `LICENSE` - **Changelog** (documentation): All notable changes to Agentic SWMM Workflow are documented here. Evidence: `CHANGELOG.md` - **Agent NL → SWMM End-to-End Evidence 2026-05-14** (documentation): Agent NL → SWMM End-to-End Evidence 2026-05-14 Evidence: `docs/agent-nl-swmm-evidence-20260514.md` - **Agent NL - SWMM End-to-End Evidence with gpt-5.5 2026-05-15** (documentation): Agent NL - SWMM End-to-End Evidence with gpt-5.5 2026-05-15 Evidence: `docs/agent-nl-swmm-gpt55-evidence-20260515.md` - The remaining 20 evidence entries are in `AI_CONTEXT_PACK.json` or `EVIDENCE_INDEX.json`. ## Rules the Host AI Must Follow - **Treat this asset as pre-work context, not a runtime environment.**: The AI Context Pack contains only an evidence-backed understanding of the project, not the project's executable state. Evidence: `docs/framework-validation/saanich-b7-network-routed-20260513/README.md`, `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/README.md`, `docs/framework-validation/saanich-cold-start-cecelia-20260513/README.md` - **When answering the user, distinguish what can be previewed from what can only be verified after install.**: The consumer value of the pre-install experience comes from reducing bad installs and misjudgments, not from pretending to be a real run. Evidence: `docs/framework-validation/saanich-b7-network-routed-20260513/README.md`, `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/README.md`, `docs/framework-validation/saanich-cold-start-cecelia-20260513/README.md` ## Questions the User Should Answer First - Which host AI or local environment do you plan to use it in? - Do you just want to experience the workflow first, or are you ready to actually install? - What matters most to you: install cost, output quality, or conflicts with your existing rules? ## Acceptance Checks - Every capability claim can be traced back to a file path in evidence_refs. - AI_CONTEXT_PACK.md does not package previews as a real run. - The user can understand who it fits, what it can do, how to start, and the risk boundaries within 3 minutes. --- ## Doramagic Context Augmentation The following sections strengthen the repository context for a host AI. Human Manual data is a reading route, and pitfall notes become operating constraints. ## Human Manual Outline Usage rule: this is only a reading route and salience signal, not factual authority. Concrete claims must still return to repo evidence or Claim Graph. Host AI hard rules: - Do not treat page titles, section order, summaries, or importance values as factual project evidence. - When explaining the Human Manual outline, state that it is only a reading route or salience signal. - Capability, installation, compatibility, runtime state, and risk claims must cite repo evidence, source paths, or Claim Graph. - **Project Overview and Goals**: importance `high` - source_paths: README.md, CITATION.cff, CHANGELOG.md - **Agent Runtime and Orchestration**: importance `high` - source_paths: agentic_swmm/agent/runtime.py, agentic_swmm/agent/repl.py, agentic_swmm/agent/planner.py, agentic_swmm/agent/intent_classifier.py, agentic_swmm/agent/executor.py - **MCP Servers and Skill Layer**: importance `high` - source_paths: integrations/mcp/README.md, integrations/skills/README.md, skills/swmm-end-to-end/SKILL.md, mcp/swmm-runner/server.js, mcp/swmm-builder/server.js - **SWMM Build, Run, and Network Synthesis**: importance `high` - source_paths: agentic_swmm/agent/swmm_runtime/run_artifacts.py, agentic_swmm/agent/swmm_runtime/run_manifests.py, agentic_swmm/agent/swmm_runtime/preflight.py, agentic_swmm/agent/swmm_runtime/postflight.py, agentic_swmm/agent/swmm_runtime/inp_parsing.py - **Calibration, Uncertainty, and Water Quality**: importance `medium` - source_paths: agentic_swmm/agent/calibration_batch.py, agentic_swmm/agent/swmm_runtime/calibration_runner.py, agentic_swmm/agent/swmm_runtime/uncertainty_plan.py, agentic_swmm/agent/swmm_runtime/design_storm.py, agentic_swmm/agent/tool_handlers/swmm_calibration.py - **GIS, Climate, Plot, Report, and Review Skills**: importance `medium` - source_paths: skills/swmm-gis/SKILL.md, skills/swmm-params/SKILL.md, skills/swmm-climate/SKILL.md, skills/swmm-plot/SKILL.md, skills/swmm-report/SKILL.md - **Modelling Memory and Learning Layer**: importance `high` - source_paths: agentic_swmm/memory/jsonl_store.py, agentic_swmm/memory/session_db.py, agentic_swmm/memory/session_repair.py, agentic_swmm/memory/recall.py, agentic_swmm/memory/parametric_memory.py - **Audit, Provenance, and Verification**: importance `high` - source_paths: agentic_swmm/audit/provenance_v1_2.py, agentic_swmm/audit/run_folder_layout.py, agentic_swmm/audit/chat_note.py, agentic_swmm/audit/llm_calls.py, agentic_swmm/agent/swmm_runtime/compare.py ## Repo Inspection Evidence - repo_clone_verified: true - repo_inspection_verified: true - repo_commit: `33306a9fee7fc4c8b0a773c7cd184a247ee5c9c8` - inspected_files: `Dockerfile`, `README.md`, `pyproject.toml`, `requirements.txt`, `docs/agent-nl-swmm-evidence-20260514.md`, `docs/agent-nl-swmm-gpt55-evidence-20260515.md`, `docs/agent-nl-swmm-gpt55-llm-engaged-evidence-20260515.md`, `docs/agent-nl-swmm-gpt55-plot-evidence-20260515.md`, `docs/agentic-runtime-ux-prd.md`, `docs/api-key-configuration.md`, `docs/byte-identical-reproducibility.md`, `docs/codex-runtime.md`, `docs/experiment-audit-framework.md`, `docs/framework-validation/BACKLOG.md`, `docs/framework-validation/saanich-b7-network-routed-20260513/README.md`, `docs/framework-validation/saanich-b7-network-routed-20260513/network.json`, `docs/framework-validation/saanich-b7-network-routed-20260513/runner_manifest.json`, `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/README.md`, `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/network.json`, `docs/framework-validation/saanich-b8-end-to-end-out1-flowing-20260513/runner_manifest.json` Host AI hard rules: - Without repo_clone_verified=true, do not claim that the source code has been read. - Without repo_inspection_verified=true, do not write README, docs, or package-file conclusions as facts. - Without quick_start_verified=true, do not claim that the Quick Start path has run successfully. ## Doramagic Pitfall Constraints These rules come from Doramagic discovery, validation, or compilation findings. The host AI must treat them as operating constraints, not background notes. ### Constraint 1: Capability evidence risk requires verification - Trigger: README/documentation is current enough for a first validation pass. - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: capability.assumptions | https://github.com/Zhonghao1995/agentic-swmm-workflow - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 2: Security or permission risk requires verification - Trigger: no_demo - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: downstream_validation.risk_items | https://github.com/Zhonghao1995/agentic-swmm-workflow - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 3: Security or permission risk requires verification - Trigger: no_demo - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: risks.scoring_risks | https://github.com/Zhonghao1995/agentic-swmm-workflow - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 4: Maintenance risk requires verification - Trigger: issue_or_pr_quality=unknown。 - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: evidence.maintainer_signals | https://github.com/Zhonghao1995/agentic-swmm-workflow - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it. ### Constraint 5: Maintenance risk requires verification - Trigger: release_recency=unknown。 - Host AI rule: Reproduce the official install and quickstart path in an isolated environment. - Why it matters: May increase setup, validation, or first-run risk for the user. - Evidence: evidence.maintainer_signals | https://github.com/Zhonghao1995/agentic-swmm-workflow - Hard boundary: Do not present this pitfall as solved, verified, or ignorable unless later evidence explicitly closes it.