"Routes agents across Qiskit's circuit, transpiler, primitives,
Scanned 9/9/2026
Install to Claude Code
npx -y skills add VectorSpaceLab/AREX-Skill --skill qiskit --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Qiskit?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vectorspacelab-qiskit)More formats (shields.io, HTML) on the badges page.
---
name: qiskit
description: "Routes agents across Qiskit's circuit, transpiler, primitives,
serialization, quantum-info, visualization, provider, and C-API workflows."
disable-model-invocation: true
metadata:
disco-role: operating
license: Apache 2.0
---
# Qiskit
Use this repo skill when a task involves the Python package `qiskit`: building quantum circuits, transpiling to backend targets, running reference primitives or basic simulators, importing/exporting OpenQASM or QPY, using quantum-information objects, drawing circuits or results, working with provider interfaces, or consuming Qiskit's public C API.
## Start here
- Read `references/module-map.md` for the package modules, public entry points, and route boundaries.
- Read `references/installation.md` before choosing extras such as `qasm3-import`, `visualization`, `crosstalk-pass`, `csp-layout-pass`, or `qpy-compat`.
- Read `references/troubleshooting.md` for mixed-package imports, missing Rust-backed `_accelerate`, optional-dependency failures, QPY compatibility, and visualization system-tool issues.
- Read `references/repo-provenance.md` before deciding whether this skill is stale for a newer Qiskit checkout or release.
- Run `python scripts/check_qiskit_environment.py --help` to see source-free smoke checks for the active Python environment.
## Minimal install and smoke check
For normal package use:
```bash
python -m pip install qiskit
python -c 'import qiskit; print(qiskit.__version__)'
```
Add targeted extras only for the workflows you need:
```bash
python -m pip install 'qiskit[qasm3-import]'
python -m pip install 'qiskit[visualization]'
```
Then run the bundled smoke helper from this skill directory:
```bash
python scripts/check_qiskit_environment.py --sections core transpiler primitives serialization quantum-info
```
## Route by task
| User task | Read |
| --- | --- |
| Build, edit, compose, parameterize, or inspect `QuantumCircuit` objects, registers, bits, gates, control-flow operations, or circuit-library gates | `sub-skills/circuit/SKILL.md` |
| Compile circuits to a backend, `Target`, basis gate set, or coupling map; use preset pass managers; debug layout, routing, translation, scheduling, seeds, or pass plugins | `sub-skills/transpiler/SKILL.md` |
| Use `StatevectorSampler`, `StatevectorEstimator`, PUBs, `PrimitiveResult`, `DataBin`, `BitArray`, expectation values, shots, precision, or primitive result access patterns | `sub-skills/primitives/SKILL.md` |
| Import/export OpenQASM 2, OpenQASM 3, or QPY; choose between text interchange and full-fidelity binary serialization; handle parser/exporter/version errors | `sub-skills/serialization/SKILL.md` |
| Work with `Statevector`, `DensityMatrix`, `Operator`, `SparsePauliOp`, `Pauli`, Clifford, random operators/states/channels, fidelities, entropy, or matrix predicates | `sub-skills/quantum-info/SKILL.md` |
| Draw circuits, histograms, distributions, states, backend maps, timelines, pass managers, or diagnose `matplotlib`, Graphviz, Pillow, pylatexenc, seaborn, or LaTeX output | `sub-skills/visualization/SKILL.md` |
| Use or implement backend/provider abstractions, `BackendV2`, `Options`, `Job`, `BasicProvider`, `BasicSimulator`, or `GenericBackendV2` fake backends | `sub-skills/providers/SKILL.md` |
| Locate Qiskit C headers/library, use `qiskit.capi` ctypes bindings, build or consume the public C API, or diagnose Rust/C-extension build failures | `sub-skills/c-api/SKILL.md` |
## Common routing decisions
- Start with `circuit` before `transpiler` when the user has not yet built a valid abstract circuit.
- Start with `transpiler` when the problem involves basis gates, coupling maps, `Target`, `Layout`, `PassManager`, or hardware constraints, even if a fake provider appears in the example.
- Start with `primitives` for sampler/estimator jobs and result containers; start with `providers` for backend classes, `BasicSimulator.run`, `GenericBackendV2`, or provider implementation surfaces.
- Start with `serialization` for QASM/QPY even when the circuit construction is also involved; route back to `circuit` only if the circuit itself is invalid before serialization.
- Start with `quantum-info` when the task is mathematical analysis of states, operators, observables, channels, or fidelities rather than execution on a backend.
- Start with `visualization` whenever output format, style, image generation, or optional drawing dependencies are the main concern.
- Start with `c-api` for `qiskit.capi.get_include()`, `get_lib()`, native headers, ctypes functions, or standalone C-library build questions.
## Bundled helper
`scripts/check_qiskit_environment.py` performs quick, source-free checks for imports, a small circuit, transpilation with `GenericBackendV2`, `StatevectorSampler`/`StatevectorEstimator`, QASM/QPY round-trips, quantum-info objects, visualization, providers, and C-API paths. Use `--sections` to narrow the checks and `--json` for machine-readable output.
## What this skill does not cover
- It does not replace domain packages such as Qiskit Aer, Qiskit IBM Runtime, Qiskit Nature, or hardware-vendor provider packages when their package-specific APIs are the main subject.
- It does not teach quantum-computing theory beyond the Qiskit object and workflow contracts needed to complete software tasks.
- It does not require the original Qiskit source checkout for runtime use; all necessary operating guidance is distilled into this skill tree.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!