# Matplotlib

> Matplotlib data visualization library. Use for plotting.

- Skill: `g1joshi/matplotlib` (Agent Skill)
- Install (CLI): `npx skillmds@latest add g1joshi/matplotlib`
- Raw SKILL.md: https://api.skillmd.com/api/skills/g1joshi/matplotlib/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: G1Joshi (https://skillmd.com/u/g1joshi)
- Updated: 2026-09-10
- Page: https://skillmd.com/skills/g1joshi/matplotlib

---


# Matplotlib

Matplotlib is the grandfather of Python plotting. It is verbose but provides **infinite control**.

## When to Use

- **Publication**: Creating figures for papers (PDF/SVG).
- **Customization**: When you need to control every pixel.
- **Backend**: It powers Seaborn and Pandas plotting.

## Core Concepts

### Figure & Axes

The container (`fig`) and the plot area (`ax`). Always use the OO interface, not `plt.plot`.

### Backends

Interactive (`Qt`, `WebAgg`) vs Static (`Agg`).

## Best Practices (2025)

**Do**:

- **Use `plt.subplots()`**: The Object-Oriented style.
- **Use `style.use('seaborn-v0_8')`**: Make it look decent by default.

**Don't**:

- **Don't use `pylab`**: It is deprecated and pollutes namespaces.

## References

- [Matplotlib Documentation](https://matplotlib.org/)