biorouter-skills

Plan and review computational fluid dynamics workflows with meshes, boundary conditions, turbulence models, OpenFOAM/FluidSim-style solvers, validation cases, convergence checks, and reproducible simulation artifacts. Use when users mention CFD, fluid dynamics, Navier-Stokes, meshing, turbulence, OpenFOAM, or flow simulation.

Run evaluator-driven hypothesis trees for improving research artifacts, prompts, analyses, and designs. Use when a task needs branch-and-prune exploration, rubric scoring, iterative refinement, or explicit comparison of competing scientific/workflow alternatives.

Detect repeated user workflows and draft native BioRouter skills, hooks, or checklist updates from observed work patterns. Use when users ask to turn repeated work, screen-history patterns, project routines, or recurring guardrails into reusable BioRouter skills.

Plan and audit BIDS neuroimaging datasets, derivatives, validation, metadata, and reproducible preprocessing. Use when users mention BIDS, fMRI, MRI, EEG/MEG/iEEG, DICOM conversion, dataset_description.json, participants.tsv, or neuroimaging derivatives.

Plan and review clinical AI modeling workflows with EHR cohorts, PyHealth-style tasks, validation, bias checks, calibration, safety boundaries, and no-PHI handling. Use when users ask about clinical prediction models, risk scores, treatment-response ML, medical reports, or clinical model evaluation.

Plan and review general time-series forecasting with classical models, ML, foundation models, backtesting, leakage checks, and uncertainty. Use for non-omics forecasting, sensor streams, operations, finance, clinical vitals, demand, or irregular temporal data.

Find and evaluate Hugging Face scientific models, datasets, Spaces, and papers with license, provenance, benchmark, and fit checks. Use when users ask for scientific ML resources, model cards, datasets, embeddings, checkpoints, or reproducible AI assets from Hugging Face.

Plan and review molecular dynamics simulations and trajectory analysis with OpenMM, GROMACS, AmberTools, MDAnalysis, MDTraj, and reproducibility checks. Use when users ask about MD setup, force fields, equilibration, production runs, trajectories, RMSD/RMSF, or binding/stability dynamics.

Plan CATE, causal forests, meta-learners, policy learning, conformal causal inference, fairness audits, subgroup stability, and heterogeneous-effect reporting. Use when causal effects vary across people, places, time, or policy rules.

Route empirical econ/finance data-source work across WRDS, CRSP, Compustat, SEC/EDGAR, NBER, SSRN, OpenAlex, Unpaywall, and common public finance/econ datasets. Use when the blocker is finding, joining, licensing, or documenting source data.

Plan social-science paper submission, journal targeting, referee reports, response letters, revise-and-resubmit execution, house style conversion, and robustness appendix organization. Use when moving an empirical paper toward submission or revision.

Plan measurement workflows: scale reliability, CFA/SEM, mediation/moderation, latent constructs, invariance, item analysis, and APA-style reporting. Use for surveys, instruments, behavioral measures, or psychometric validation.

Plan quantitative text-as-data workflows: dictionaries, topic models, embeddings, sentiment, supervised coding, LLM-assisted labels, validation, and measurement-error checks. Use for corpora, speeches, filings, interviews, open text, or policy documents.

Plan, review, and package Stata empirical research workflows: .do files, data audit, logs, graph export, table building, replication, linting, and referee-response updates. Use when a user works in Stata or asks for .do/.dta research workflows.

Route structural economics and theory work: equilibrium models, game theory, DSGE/HANK, IO demand/supply, identification proofs, calibration, simulation, and estimation design. Use when reduced-form causal tools are not enough.

Guide complex survey analysis with strata, PSUs, replicate weights, domain analysis, calibration, weighted GLMs, and design-aware inference. Use for CPS, ACS, NHANES, survey microdata, polling, or any weighted sample design.

Review clinical decision support, treatment-plan, and clinical-report workflows for evidence grounding, patient-safety language, scope limits, and no-PHI handling. Use when a task asks for CDS, care pathways, treatment plans, or clinical report drafting.

Plan drug-discovery benchmark workflows with TDC-style datasets, ADMET, virtual screening, molecular property prediction, scaffold splits, leakage checks, and assay provenance. Use when evaluating compound models, benchmarks, or drug-discovery ML results.

Plan empirical finance/accounting workflows with Compustat/CRSP-style panels, event studies, abnormal returns, clustering, fixed effects, disclosure measures, and table replication. Use for accounting, finance, corporate, or market-data empirical research.

Plan and review GPU/scientific-compute optimization across PyTorch, CUDA-aware batching, memory pressure, mixed precision, profiling, and reproducibility. Use when workflows are slow, GPU-bound, memory-bound, or need scalable training/inference plans.

Structure research grants, specific aims, significance/innovation/approach, reviewer-risk responses, milestones, and funder-fit checks. Use when users ask for grant planning, NIH-style aims, fellowship statements, or funding strategy.

Guide constraint-based metabolic modeling with COBRApy, flux balance analysis, gene-protein-reaction rules, media constraints, knockouts, objective functions, and model provenance. Use when users ask about genome-scale metabolic models or flux predictions.

Route astronomy, physics, chemistry, and numerical-science workflows with Astropy, QuTiP-style simulations, units, coordinate systems, numerical solvers, and reproducible physical constants. Use when requests mention astronomy, physics, physical chemistry, units, coordinate transforms, spectra, or numerical experiments.

Plan qualitative coding, thematic analysis, codebooks, memoing, inter-rater checks, audit trails, and mixed-methods evidence synthesis. Use when data are interviews, free text, field notes, survey responses, or qualitative documents.

Review scientific software, medical-device, and lab-process work against ISO 13485-style quality systems, design controls, validation evidence, traceability, and risk management. Use when tasks mention regulatory readiness, quality management, validation, design controls, or audit trails.

Evaluate papers, scholars, venues, impact claims, publication fit, reviewer expectations, and research positioning without overclaiming bibliometrics. Use when users ask where to submit, how strong a paper is, or how to evaluate a scholar/topic.

Plan scientific schematics, infographics, posters, slides, figure panels, Mermaid diagrams, and visual abstracts with source-to-visual traceability. Use when users need a visual explanation, figure plan, poster, slide deck, schematic, or graphical abstract.

Plan simulation and optimization workflows using SimPy, PyMOO, FluidSim-style models, discrete-event simulation, multi-objective optimization, sensitivity analysis, and validation against known cases. Use when a scientific problem asks for simulation, optimization, scheduling, resource allocation, design search, or what-if modeling.

Plan biomedical imaging workflows across DICOM, pathology slides, PACS metadata, PathML/histolab-style tiling, image QC, cohort metadata, and privacy-safe reporting.

Use Zotero, DOI/PMID lookup, Crossref/OpenAlex, MarkItDown, and citation-file hygiene to build, clean, and verify reference libraries and bibliographies. Use when the user asks for Zotero, BibTeX, RIS, CSL, citation cleanup, or source-library deduplication.

Plan applied econometrics workflows in Python, R, or Stata: panel data, fixed effects, IV, DiD/event studies, regression discontinuity, clustering, marginal effects, and table replication. Use for social-science, finance, accounting, public-policy, or economics empirical analysis.

Handle geospatial research workflows with GeoPandas, raster/vector data, coordinate reference systems, spatial joins, maps, and spatial statistics. Use when data has locations, regions, polygons, rasters, tracks, or environmental/geographic covariates.

Plan safe lab-automation workflows for Opentrons, cloud labs, Ginkgo-style APIs, plate maps, run metadata, and protocol simulation. Use when a request involves robots, wet-lab run planning, samples, plates, liquid handling, or cloud-lab execution.

Guide computational materials and molecular-structure workflows with pymatgen, ASE-style structures, crystals, compositions, phase diagrams, descriptors, and provenance for materials datasets.

Plan analysis for biosignals and neural recordings: ECG/EDA/PPG with NeuroKit2, spike sorting/Neuropixels metadata, peri-event analysis, quality control, and reproducible signal-processing pipelines.

Route and review quantum/scientific simulation work using Qiskit, Cirq, PennyLane, QuTiP, variational circuits, quantum chemistry, and simulation validation. Use when a task mentions quantum circuits, quantum ML, Hamiltonians, or open quantum systems.

Plan and review scalable scientific data workflows with Polars, Dask, Zarr, TileDB, Arrow/Parquet, HDF5, and dataset versioning. Use when scientific data is too large for naive pandas/R data frames or when the user asks about chunking, columnar storage, lazy execution, cloud/object storage, or reproducible data pipelines.

Plan scientific ML workflows with scikit-learn, PyTorch Lightning, transformers, graph neural networks, active learning, SHAP/explainability, and leakage checks. Use when applying ML to biomedical, materials, chemistry, imaging, or other scientific datasets.

Design and critique statistical models with statsmodels, PyMC, survival models, hierarchical models, power analysis, posterior predictive checks, and uncertainty reporting. Use when a request needs principled statistical modeling rather than a generic machine-learning fit.

Run systematic-review workflows: protocol/PICO framing, search strings, screening logs, PRISMA flow, extraction tables, risk-of-bias checks, and evidence synthesis with citation verification.