cirq — Claude Code skill: quantum circuits in Python
A skill for Google Quantum AI's Cirq framework: build and simulate quantum circuits, model noise, run on hardware. Useful when your Python project touches quantum computing.
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a claude-codeWhat cirq does
The skill hands your agent reference material on Cirq: circuit building, simulation, transformations, hardware integration and noise modeling. It ships references/building.md, references/simulation.md, references/transformation.md, references/hardware.md and references/noise.md, plus code examples. The agent reads them with Read, edits files with Write and Edit, and runs code through Bash.
Examples target Cirq 1.7.0 on Python 3.11+. Locally, uv pip install "cirq-core==1.7.0" is enough; OpenQASM import also needs ply. Keep every cirq-* package in one environment on the same release, because cirq-google, cirq-ionq, cirq-aqt and cirq-pasqal ship in step with the core. Azure is the exception: it runs its own Cirq 1.6.1 stack, and forcing 1.7.0 into it breaks the setup.
Cloud execution needs provider credentials, network access and assigned targets. Google Quantum Engine requires an approved GCP project, and bundled virtual processor names prove nothing about live access. The skill covers Google, IonQ, Azure Quantum, AQT and Pasqal. For IBM hardware you want qiskit, for quantum ML with autodiff pennylane, for physics simulations qutip.
What it helps with
- build a parameterized circuit and sweep it over a range of values
- model noise and see how it shifts the measurement results
- compile a circuit down to a specific processor's gateset
- submit a job to Google Quantum AI or a partner backend
- work through transformers and qubit routing
Example prompts for the agent
- Build a Bell state circuit on two qubits and run 1000 shots on
cirq.Simulator. - Sweep the
thetaparameter from 0 to 2π and show me the histograms. - Add a depolarizing noise model and compare against the ideal simulation.
How to install cirq
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a claude-code2Manually: put the files in place
~/.claude/skills/cirq/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/cirq/
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a codex2Manually: put the files in place
.agents/skills/cirq/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a cursor2Manually: put the files in place
.cursor/skills/cirq/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install K-Dense-AI/scientific-agent-skills cirq2Manually: put the files in place
.github/skills/cirq/SKILL.md1Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a gemini-cli2Platform CLI
gemini skills install https://github.com/K-Dense-AI/scientific-agent-skills --path skills/cirqManually: put the files in place
.devin/skills/cirq/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about cirq
How do I install cirq in Claude Code?
Run npx skills add K-Dense-AI/scientific-agent-skills --skill cirq -a claude-code. The skill lands in ~/.claude/skills/cirq/SKILL.md.
What does the cirq skill do?
It gives the agent Cirq references and examples: circuit construction, simulation, transformations, hardware integration and noise modeling.
What do I need to run circuits on real hardware?
Provider credentials, network access and assigned targets. Google Quantum Engine also needs an approved GCP project.
