All skills

Drafts competitive funding proposals tailored to NSF, NIH, DOE, DARPA, and Taiwan NSTC review criteria and formatting rules.

Zero-shot univariate time series forecasting with Google's TimesFM foundation model, including calibrated prediction intervals.

Generates correct, up-to-date seaborn 0.13 code for distribution, relationship, and categorical plots from pandas DataFrames.

Guides Claude to write correct, up-to-date scikit-bio code for sequence analysis, alignment, phylogenetics, and microbiome diversity statistics.

A skill that guides rigorous statistical modeling in Python with statsmodels — regression, GLM, discrete choice, and time series, plus diagnostics and inference.

Helps Claude write correct Zarr 3.x code for chunked, compressed N-dimensional arrays on local disk or cloud object storage.

Lets Claude query the key-free U.S. Treasury Fiscal Data REST API for national debt, auctions, exchange rates, and federal revenue/spending.

An end-to-end guide for research statistics: picking the right test, verifying assumptions, reporting effect sizes, and writing APA-style results.

Runs evidence-aware research ideation with independent generation, transparent scoring, adversarial review, and an auditable decision log.

A skill that guides Claude to use Vaex for lazy, out-of-core analysis, visualization, and ML on billion-row tabular datasets that exceed RAM.

Guides Claude to perform fast genomic interval operations and bioinformatics file I/O (BED/VCF/BAM/GFF) on Polars DataFrames with polars-bio.

A skill that guides classical machine learning in Python with scikit-learn — classification, regression, clustering, preprocessing, evaluation and tuning — following leakage-free best practices.

Prepares journal manuscripts, conference papers, posters, and grant documents by verifying current official venue rules first, then using bundled LaTeX scaffolds.

A skill that guides Claude in building, analyzing, and visualizing networks and graphs with Python's NetworkX.

Builds conference-ready research posters in beamerposter, tikzposter, or baposter — from layout to print-ready PDF.

Extract layout-preserved text with per-item bounding boxes, run OCR, and render page PNGs from PDFs, Office files, and images — all fully local.

A skill that steers Claude toward idiomatic Matplotlib code for publication-quality figures.

Turn a plain-language description into a publication-quality infographic, with AI quality scoring that only re-renders when the result falls below a threshold.

Guides you through UMAP embeddings, parameter tuning, and clustering preprocessing with ready-to-run scikit-learn code.

Conservatively detects the CPU, memory, disk and accelerator limits your current process actually has, emitting a redacted JSON snapshot plus safe worker-planning helpers.