Matplotlib publication plot starter
Skill ma-compbio-lab/SkillFoundry/skills/visualization-and-reporting/matplotlib-publication-plot-starter
A framework for discovering, compiling, and validating reusable skills for scientific agents.
npx -y skills add ma-compbio-lab/SkillFoundry --skill matplotlib-publication-plot-starterAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
1.2 KB, 270 tokens by cl100k_base, as published. Nobody here has run it
Matplotlib Publication Plot Starter
Use this skill to turn a small tabular dataset into a deterministic publication-style figure and compact fit summary.
What it does
- Loads a toy TSV of
xandyobservations. - Fits a simple linear trend with NumPy.
- Renders a two-panel Matplotlib figure with the fitted line and residual plot.
- Saves a PNG plus a JSON summary with fit statistics.
When to use it
- You need a local starter for publication plots.
- You want a reproducible figure-generation pattern that can later be swapped for real data.
Example
slurm/envs/statistics/bin/python skills/visualization-and-reporting/matplotlib-publication-plot-starter/scripts/render_publication_plot.py \
--input skills/visualization-and-reporting/matplotlib-publication-plot-starter/examples/toy_measurements.tsv \
--png-out scratch/matplotlib/publication_plot.png \
--summary-out scratch/matplotlib/publication_plot_summary.json
Verification
- Skill-local tests:
python3 -m unittest discover -s skills/visualization-and-reporting/matplotlib-publication-plot-starter/tests -p 'test_*.py' - Repository smoke:
python3 -m unittest tests.smoke.test_cross_cutting_domain_skills -v