PennyLane Quantum Circuits
Build, train and check PennyLane quantum circuits, hybrid quantum-classical ML and VQE/QAOA workflows
- Repository
- K-Dense-AI/scientific-agent-skills
- Stars
- 55,780
- Licence
- MIT
- Files
- 8 files
What it does
Once installed, the agent follows a structured PennyLane workflow for quantum computing: it designs small parameterized quantum circuits, checks them against known analytic values and gradients, trains them with classical optimizers, and validates results against classical baselines. It also covers molecular energy estimates (VQE), QAOA optimization, and hybrid PyTorch or JAX models, and explains how to move validated circuits to hardware provider plugins. It is a guide to doing this correctly, not a quantum computer itself; the bundled examples use small synthetic problems.
Great for
- Researchers or students learning variational quantum algorithms with PennyLane
- Developers prototyping hybrid quantum-classical machine learning models
- Anyone who needs to validate a quantum circuit on a simulator before trying real hardware
Skip it if
- Hardware-specific compilation or transpilation, which the skill says should use qiskit or cirq instead; using it here could give wrong results for that task
- Claims of quantum advantage or chemical accuracy: the skill itself warns that small local examples do not establish these, so relying on them for publication-level conclusions without independent checks is risky
- Running jobs on paid quantum hardware: provider credentials and network access are needed, and the skill notes authenticated hardware jobs were not executed during its review
Try asking
- “Build a one-qubit PennyLane circuit, check its gradient against finite differences, and optimize a parameter”
- “Train a small hybrid quantum classifier with PyTorch and report held-out accuracy against a classical baseline”
- “Estimate the H2 molecule ground-state energy with a UCCSD ansatz and check particle number”
- “Solve a small MaxCut instance with QAOA and compare against enumerated solutions”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 8 original files, the MIT licence and an attribution note.
pennylane.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip pennylane.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip pennylane.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Use a dedicated virtual environment as the installation section recommends, since PennyLane pins NumPy 2+ and specific JAX versions
- Ask the agent to report seeds, package versions and uncertainty with every result, and to validate against a classical reference
- Keep local simulation for development; only add credentials and network access when you deliberately target real hardware
What's inside
- references/advanced_features.md
- references/devices_backends.md
- references/getting_started.md
- references/optimization.md
- references/quantum_chemistry.md
- references/quantum_circuits.md
- references/quantum_ml.md
- SKILL.md
Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.