FluidSim CFD Simulation Guardrails

A rigorous workflow skill for planning, validating, restarting, and analyzing FluidSim 0.9 pseudospectral CFD runs with explicit numerical and HPC safety gates.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

Guides solver selection across FluidSim 0.9.0 (ns2d, ns3d, Boussinesq and stratified variants) and forces inspection of generated default parameters before editing.
Builds a strict JSON plan with hard bounds on CPU, RAM, disk, wall-time, CFL, resolution and dealiasing, then checks it with bundled solver_config_validator.py and grid_resource_estimator.py.
Enforces the order: dry-run script → tiny serial pilot → independent grid and timestep refinement → only then a site-specific MPI job, and never auto-submits MPI.
Ships read-only analysis tools for output inventory, energy budgets and restart compatibility, and refuses the upstream --modify-params option that executes arbitrary Python.
Explicitly rejects "it finished and the plot looks smooth" as evidence of convergence, demanding spectra, CFL history, and conservation residuals.

Who it's for

Graduate students and researchers doing periodic pseudospectral CFD.
Computational scientists wiring FFTW/PFFT/P3DFFT backends on HPC clusters.
Teams that must keep reproducible provenance: lockfiles, checksums, restart lineage.

Examples

"Run a forced 2D turbulence ns2d case at 512²" — the skill first pins equations, units and acceptance criteria, then emits a validated config plus a resource estimate.
"Can I trust the results in this run directory?" — output_inventory.py and budget_summary.py audit files and energy-budget residuals.
"Restart from the last checkpoint with tweaked forcing" — restart_compatibility.py compares solver, grid, state variables and versions before anything runs.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/fluidsim/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)

Download the ZIP with the button below.
In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.

↓ Download ZIP

Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/fluidsim folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/fluidsim/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/fluidsim ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

Open a terminal and create the skills folder: mkdir -p ~/.claude/skills
Clone the repository: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
Copy just this skill: cp -r scientific-agent-skills/skills/fluidsim ~/.claude/skills/
Confirm Python 3.11+ is available: python3 --version
For real simulations, create an isolated env: uv venv --python 3.11 && uv pip install "fluidsim[fft]==0.9.0" "fluidfft==0.4.5" "pyFFTW==0.15.1"
Restart Claude Code and prompt something like "review my ns2d FluidSim configuration" to trigger the skill.
For MPI or native FFT backends, have your cluster admin install a compatible MPI runtime and FFTW development headers first.

View source on GitHub ↗License: MIT