Computational Scientific Machine Learning Lab (CSML)
Official@csml-rpi · United States of America
A cross‑disciplinary research group working on transformative research to solve challenging problems in aerospace engineering.
Agent Skills by Computational Scientific Machine Learning Lab (CSML)
Showing 17 vetted skills indexed across 1 GitHub repositories.
csml-rpicfd-pipeline
Automate CFD research pipelines from literature review to publication.
csml-rpicfd-experiment
Automate OpenFOAM CFD case execution with validation and retry logic.
csml-rpicfd-hypothesis
Generate multiple CFD hypotheses with solver, geometry, and constraint specifications.
csml-rpicfd-mesh-gate
Automate mesh refinement studies and lock optimal mesh parameters for CFD simulations.
csml-rpicfd-paper
Automates CFD research paper writing including LaTeX drafting and iterative reviews.
csml-rpicfd-open-discovery
Automate CFD model proposal, modification, simulation, and multi-metric scoring.
csml-rpicfd-code-modify
Implement custom turbulence, viscosity, or source models in OpenFOAM cases.
csml-rpicfd-requirements
Convert CFD hypotheses into FoamAgent requirement JSON files.
csml-rpicfd-analyze
Aggregate CFD case data and calculate trends and correlations among key metrics.
csml-rpicfd-literature
Retrieve and normalize CFD research papers from Semantic Scholar, OpenAlex, and arXiv into structured JSON.
csml-rpicfd-viz
Generate CFD visualization figures with PyVista and vision-based QA.
csml-rpicfd-interpret
Analyze CFD diagnostic images and logs to validate simulation results.
csml-rpicfd-research
Automate CFD research workflows from literature review to result analysis.
csml-rpicfd-orchestrator
Route CFD research requests into appropriate simulation, analysis, or documentation pipelines.
csml-rpicfd-foamagent-runtime
Automate Foam-Agent CFD simulation workflows from planning through analysis.
csml-rpicfd-code-mod
Automate implementing, modifying, and testing custom OpenFOAM models in CFD cases.
csml-rpicfd-mesh-independence
Conduct mesh sensitivity studies and regression analysis for CFD simulations.
Frequently Asked Questions About Computational Scientific Machine Learning Lab (CSML)
FAQPage SchemaWhat specific research tasks are enabled by these capabilities?▼
These capabilities enable end-to-end research cycles, including literature normalization, hypothesis generation, OpenFOAM case execution, mesh refinement studies, and automated LaTeX drafting for publication. Users can manage complex simulation parameters, perform diagnostic image interpretation, and conduct regression analysis on multi-metric datasets derived from fluid dynamics experiments.
Which personas benefit from these research capabilities?▼
Aerospace engineers, computational fluid dynamics researchers, and scientific investigators benefit from these capabilities. The system is designed for professionals managing high-fidelity simulation environments who require rigorous validation, mesh independence verification, and structured documentation of complex physical models within the OpenFOAM ecosystem.
What are the primary dependencies for running these simulation environments?▼
The environment requires a functional OpenFOAM installation for solver execution and PyVista for visualization tasks. Users must provide structured JSON requirement files to define hypotheses, geometry constraints, and turbulence model modifications, which are then processed through the orchestrator to manage simulation runtime and data aggregation.