Scientific Visualization Tools
Designscientific-visualization-toolsScientific Visualization Tools is an Agent Skill for design published in the DrugClaw/DrugClaw repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.
Scientific visualization workflow guide for publication-ready static figures with seaborn or matplotlib and interactive figures with Plotly. Use when the user asks for scientific plots, cohort or assay figures, publication graphics, dashboards, or reusable plotting scripts for research datasets.
How to use this skill
Bring this guide into your coding agent with a prompt tailored to the tool you use.
- Open your project in Codex.
- Copy and paste the prompt below.
- Review files and risks before approving any action.
Inspect the Agent Skill at https://github.com/DrugClaw/DrugClaw/blob/HEAD/skills/science/scientific-visualization-tools/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/scientific-visualization-tools, but wait for my approval before writing files or running scripts.
Copying this prompt does not install or run the skill. Review third-party files before use.
Scientific Visualization Tools
Use this skill when the user needs a figure artifact rather than only a numeric summary.
Typical triggers:
- publication-ready scatter, box, violin, bar, or heatmap figures
- interactive HTML charts for exploratory research data review
- small reusable plotting scripts for assay, omics, or cohort tables
- consistent styling across scientific plots
Environment Check
which python3 || true
python3 - <<'PY'
mods = ["pandas", "matplotlib", "seaborn", "plotly"]
for name in mods:
try:
__import__(name)
print(f"{name}: ok")
except Exception as exc:
print(f"{name}: missing ({exc})")
PY
If key plotting modules are missing, recommend the optional drug-sandbox image documented in docs/operations/science-runtime.md.
Bundled Assets
templates/publication_plot.pytemplates/interactive_plot.py
Preferred Workflow
- Decide first whether the output should be static publication art or interactive exploration.
- Keep the plotting script parameterized by column names rather than hardcoding one dataset.
- Save the figure and a small JSON summary of what was plotted.
- Do not use interactive charts where a paper-ready static figure is required.
- Do not claim statistical meaning from a plot unless the underlying analysis is also reported.
Static Publication Plots
python3 templates/publication_plot.py \
--input figures/assay.csv \
--kind box \
--x-column arm \
--y-column response \
--color-column arm \
--output figures/assay_box.png \
--summary figures/assay_box.json
Supported baseline kinds:
scatterlineboxviolinbarheatmap
Interactive Plotly Charts
python3 templates/interactive_plot.py \
--input figures/cohort.csv \
--kind scatter \
--x-column age \
--y-column biomarker \
--color-column response \
--output figures/cohort_scatter.html \
--summary figures/cohort_scatter.json
Use this for exploratory review, dashboards, and lightweight sharing.
Related Skills
For statistical inference behind a plot, activate stat-modeling-tools.
For Kaplan-Meier and time-to-event figures, activate survival-analysis-tools.
For broader scientific-writing and manuscript-structure work, activate scientific-workflow-tools.