pennylane

Hardware-agnostic quantum ML framework with automatic differentiation. Use when training quantum circuits via gradients, building hybrid quantum-classical models, or needing device portability across IBM/Google/Rigetti/IonQ. Best for variational algorithms (VQE, QAOA), quantum neural networks, and integration with PyTorch/JAX/TensorFlow. For hardware-specific optimizations use qiskit (IBM) or cirq (Google); for open quantum systems use qutip.

1.07x

Quality

86%

Does it follow best practices?

Impact

89%

1.07x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Quality

Discovery

100%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

This is an excellent skill description that clearly defines a specific niche (hardware-agnostic quantum ML with autodiff), provides comprehensive trigger terms covering algorithms, frameworks, and hardware vendors, and explicitly delineates boundaries by naming alternative skills for adjacent use cases. The 'Use when' clause is well-structured and the negative guidance further reduces conflict risk.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions and capabilities: 'training quantum circuits via gradients', 'building hybrid quantum-classical models', 'device portability across IBM/Google/Rigetti/IonQ', 'variational algorithms (VQE, QAOA)', 'quantum neural networks', and 'integration with PyTorch/JAX/TensorFlow'.	3 / 3
Completeness	Clearly answers both 'what' (hardware-agnostic quantum ML framework with automatic differentiation) and 'when' (explicit 'Use when' clause covering training quantum circuits, building hybrid models, needing device portability). Also includes negative guidance on when NOT to use it, pointing to alternatives.	3 / 3
Trigger Term Quality	Excellent coverage of natural terms a user would mention: 'quantum ML', 'automatic differentiation', 'gradients', 'hybrid quantum-classical', 'VQE', 'QAOA', 'quantum neural networks', 'PyTorch/JAX/TensorFlow', and specific hardware vendors like 'IBM/Google/Rigetti/IonQ'. These are terms practitioners would naturally use.	3 / 3
Distinctiveness Conflict Risk	Highly distinctive with a clear niche in quantum ML with automatic differentiation and device portability. The description explicitly differentiates from related skills (qiskit for IBM-specific, cirq for Google-specific, qutip for open quantum systems), minimizing conflict risk.	3 / 3
	Total	12 / 12 Passed

Implementation

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a well-structured skill with strong actionability and excellent progressive disclosure. The main weaknesses are moderate verbosity (best practices list, resources section, plugin installation list) and missing validation/error-handling checkpoints in the multi-step workflows. The code examples are high quality and executable.

Suggestions

Trim the Best Practices section to 3-4 non-obvious, PennyLane-specific tips (e.g., remove generic advice like 'test locally' and 'start with simulators').

Add validation checkpoints to workflows — e.g., check convergence in VQE (if energy change < threshold, stop), handle hardware errors in device switching.

Remove or condense the Resources section — Claude doesn't browse URLs, and the reference files already provide detailed coverage.

Dimension	Reasoning	Score
Conciseness	The skill is reasonably efficient but includes some unnecessary content: the 'Best Practices' section has 10 items many of which are general advice Claude already knows (e.g., 'test locally', 'start with simulators'), the Resources section with external links adds little value for Claude, and the installation section listing every plugin is somewhat verbose. However, the core code examples are lean.	2 / 3
Actionability	The skill provides fully executable, copy-paste ready code examples for the quick start, variational classifier, VQE workflow, and device switching. Code includes imports, device creation, circuit definition, and optimization loops — all concrete and specific.	3 / 3
Workflow Clarity	The common workflows are clearly sequenced with numbered steps and executable code. However, there are no validation checkpoints or error recovery steps — e.g., the VQE workflow doesn't mention checking convergence criteria, and the device-switching workflow doesn't address handling hardware errors or queue failures.	2 / 3
Progressive Disclosure	Excellent progressive disclosure structure: a concise overview with quick start code, then core capabilities summarized with clear one-level-deep references to specific reference files (quantum_circuits.md, quantum_ml.md, etc.), plus a consolidated navigation section at the bottom.	3 / 3
	Total	10 / 12 Passed

Validation

90%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 10 / 11 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning

	Total	10 / 11 Passed

Repository: K-Dense-AI/claude-scientific-skills
Commit: 25e1c0f

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.