---

name: visualization description: "Guides agents drawing Qiskit circuits, histograms, distributions, states, backend maps, timelines, pass managers, and optional visualization dependencies." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Qiskit visualization workflows

Use this sub-skill when the task involves qiskit.visualization: circuit drawers, histograms, distributions, state plots, backend/device maps, pass-manager drawings, timelines, output formats, styles, or visualization optional dependencies.

Read next

• references/workflows.md for output-mode, plot, and file-saving recipes.
• references/troubleshooting.md for missing matplotlib, Graphviz, Pillow, pylatexenc, seaborn, LaTeX, and style-file failures.
• ../../references/installation.md for the visualization extra.
• ../../scripts/check_qiskit_environment.py --sections visualization for a source-free plotting smoke check.

Include here

• circuit_drawer, QuantumCircuit.draw(), output modes text, mpl, latex, and latex_source.
• Counts and distribution plots such as plot_histogram and plot_distribution.
• State visualizations such as Bloch, city, Hinton, Pauli-vector, and qsphere plots.
• Device/backend maps, circuit layout plots, pass-manager drawings, and timeline views.
• Visualization dependency, style, backend, and file-output troubleshooting.

Exclude or route elsewhere

• Building the circuit before drawing belongs in ../circuit/SKILL.md.
• Producing counts or expectation values belongs in ../primitives/SKILL.md or ../providers/SKILL.md.
• Mathematical state/operator construction belongs in ../quantum-info/SKILL.md.
• Backend compilation before layout visualization belongs in ../transpiler/SKILL.md.

Default route

Start here when the user mentions drawing, plotting, output format, images, histograms, style JSON, LaTeX, Graphviz, matplotlib, or visualizing a backend/circuit layout.

What to remember

• Qiskit's visualization module intentionally allows some user-controlled labels and external-tool calls; use trusted inputs for LaTeX and Graphviz modes.
• Text drawing is safest and dependency-light; mpl and latex modes need optional Python and system packages.
• A generated figure object often still needs savefig() or a filename= argument when the user wants an image file.