ras-end-to-end

ras-end-to-end is a skill for Claude Code, Codex from Zhonghao1995/Agentic-HEC-RAS. It costs 113 tokens per session (1,589 once invoked), scanned A, original, MIT.

A step-by-step procedure for working with HEC-RAS projects and results. HEC-RAS is software for modelling water flow and flooding in rivers and other waterways.

In plain words
What is it for?
Use it to copy projects, edit flow boundaries, run existing plans, read HDF result files, compare scenarios, create plots, and audit results.
Why use it?
It defines which steps are needed for reading results, running a model, comparing scenarios, and checking outputs. This helps prevent reporting numbers or plots before the model has passed the required checks.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to copy projects, edit flow boundaries, run existing plans, read HDF result files, compare scenarios, create plots, and audit results.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/zhonghao1995/agentic-hec-ras/ras-end-to-end
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add Zhonghao1995/Agentic-HEC-RAS --skill ras-end-to-end
Clone the repo
git clone --depth 1 https://github.com/Zhonghao1995/Agentic-HEC-RAS

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ras-end-to-end

README.md
[![agentmods](https://agentmods.dev/badge/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end/github.svg)](https://agentmods.dev/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end)
Your own site
<a href="https://agentmods.dev/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end"><img src="https://agentmods.dev/badge/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ras-end-to-end

Your own site · 80×15
<a href="https://agentmods.dev/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end"><img src="https://agentmods.dev/badge/skills/zhonghao1995/agentic-hec-ras/ras-end-to-end.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 113 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,589 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00113 $0.01589
Opus 5 $0.00056 $0.00794
Sonnet 5 $0.00023 $0.00318
Haiku 4.5 $0.00011 $0.00159

Measured 12d ago against content hash 05044b97288f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

ras-end-to-end scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ras-end-to-end/SKILL.md · 94 lines

How it starts

The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.

HEC-RAS End-to-End (SOP)

The other ras-* skills describe one tool group each. This one says how they fit together, what must be true before a number or figure is reported, and when to stop.

1. Pick an operating mode (state it in the first reply)

Mode When Chain HEC-RAS install
A. Read / plot only a .p##.hdf already exists, or no HEC-RAS install ras-results -> ras-audit none
B. Run an existing plan a .prj and a goal ("run the 2D plan") ras_project_copy -> ras-model (info) -> ras-runner -> ras-results -> ras-audit yes
C. What-if / compare "what if inflow is +20 %", "shorter time step", "before vs after" copy baseline + copy variant -> ras-model (edit variant only) -> ras-runner x2 -> ras-results (compare + plots) -> ras-audit yes
D. New event a hydrograph / hyetograph CSV to drive the model ras-model (ras_flow_set_hydrograph values) -> B yes

If the goal only needs Mode A, do not touch the model at all: it is faster and needs no HEC-RAS. Building a NEW model (terrain, 2D mesh, cross sections, infiltration layers) is out of scope: that is RAS Mapper / GUI work — say so and stop.

2. Standard run directory (create it, use it everywhere)

runs/<case>/
  00_baseline/     COPY of the whole project folder (ras_project_copy) — run in place
  00_<variant>/    second COPY for a what-if (edit + run in place)
  03_plots/        every PNG (ras_plot)
  manifest.json    ras_manifest_write, written last

Results (<name>.p##.hdf, .computeMsgs.txt, .bco##) land next to the copied .prj.

3. Hard rules

  1. Copy first, always the whole folder (ras_project_copy). Plan / flow edits rewrite .p## / .u## in place and a run writes results next to the project; terrain, land cover and DSS files are referenced relatively, so copying only the .prj breaks the run.
  2. Gate on status / usable, not ok. ras_run returns ok:true whenever the call returned. Usable = status in {completed, completed_with_warnings}; unstable / failed / no_results / unknown = do not report the numbers, surface reasons, metrics, log_tail, log_file.
  3. Discover ids before using them. ras_project_info (plans, geometries, flow files, boundary indices, river/reach/RS), ras_results_summary (cross-section labels, mesh names, reference line names). Never invent a plan number, boundary, station or mesh name.
  4. Variants are copies, not plans in the same folder. A what-if edit goes on its own copy so the baseline HDF is never overwritten (same plan number, different folder).
  5. Edit the flow file the plan actually uses. ras_flow_set_hydrograph with plan resolves the plan's .u##; several plans can share one .u## — say so if they do.
  6. Long runs block. A 2D run can take minutes to hours; set num_cores (2-8), keep the client's MCP timeout in mind, and never start a second run of the same copy while one is running.
  7. Evidence only. Report a number only from a tool return; report a figure only when ras_plot returned ok:true and png; say "not run" for anything skipped.

Read the full file on GitHub · 94 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 94 lines · 113 tokens per session scan A 05044b97288f

Subscribe to this mod's changes

ras-end-to-end is a skill published in the GitHub repository Zhonghao1995/Agentic-HEC-RAS (1 stars, last pushed 26d ago), licensed MIT. It adds 113 tokens to every session and 1,589 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.