Add praxis skill: experimental lab data analysis and publication plotting

[zmtsikriteas]

Summary

Adds Praxis, a skill for scientific data analysis and publication-quality plotting of experimental lab data. It fills a gap in the current collection: the existing 170+ skills cover bioinformatics, genomics, databases, and general-purpose libraries well, but there is no skill for analysing raw instrument data from materials science, chemistry, physics, and engineering labs.

Praxis provides:

• 21 technique-specific modules: XRD, SAXS, impedance spectroscopy (EIS), DSC/TGA, tensile testing/DMA, FTIR/Raman/UV-Vis, XPS, NMR, mass spectrometry, chromatography (GC/HPLC), BET surface area, nanoindentation, hardness testing, magnetometry (VSM/SQUID), I-V curves (solar cells, diodes), C-V measurements, dielectric analysis, piezoelectric characterisation, SEM/EDS, AFM, and thermal conductivity
• 10 general analysis modules: curve fitting (10+ models + custom equations), peak detection/deconvolution, baseline correction, smoothing, FFT/filtering, statistics, interpolation, normalisation, analysis templates, and report generation
• 16 auto-detected data formats: CSV, Excel, HDF5, MATLAB .mat, Bruker .brml, Gamry .dta, JCAMP-DX, and more
• 9 journal style files: Nature, Science, ACS, Elsevier, Wiley, RSC, Springer, IEEE, MDPI (exact column widths, fonts, DPI)
• Colourblind-safe palettes: Okabe-Ito, Tol, uchu (OKLCh-based)
• Batch processing and analysis templates (save/replay pipelines)

Structure

scientific-skills/praxis/
├── SKILL.md
├── assets/styles/          9 journal .mplstyle files
├── references/             cookbook (50+ techniques), workflows, plot types, techniques, palettes, journal styles
└── scripts/
    ├── core/               loader, plotter, exporter, utils
    ├── analysis/           fitting, peaks, baseline, smoothing, fft, stats, interpolation, normalisation, templates, report
    ├── techniques/         21 technique modules
    └── batch/              batch processing

56 files, ~15,000 lines of Python. MIT licensed.

Source

Full repo with tests, examples, and README: https://github.com/zmtsikriteas/praxis

Why this belongs here

The current skill collection is strong on bioinformatics and computational tools, but researchers doing experimental work (XRD scans, tensile tests, impedance measurements, FTIR spectra, etc.) have no skill to help them go from raw data to publication figures. Praxis closes that gap.