cirq
Google quantum computing framework. Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments. Best for Google hardware, noise modeling, and low-level circuit design. For IBM hardware use qiskit; for quantum ML with autodiff use pennylane; for physics simulations use qutip.
83
81%
Does it follow best practices?
Impact
82%
1.24xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
89%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 a strong skill description that excels at disambiguation and completeness. It clearly identifies its niche (Google quantum hardware), provides explicit 'Use when' triggers, and helpfully directs users to alternative skills for non-Google quantum tasks. The main weakness is that the specific capabilities could be more granular—listing concrete actions like 'compile circuits to Google hardware,' 'run noise simulations,' or 'perform randomized benchmarking' would strengthen specificity.
Suggestions
Add more granular concrete actions (e.g., 'compile circuits to Google hardware, simulate noise channels, perform randomized benchmarking, run on Google quantum processors') to improve specificity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (Google quantum computing) and some actions like 'designing noise-aware circuits' and 'running quantum characterization experiments,' but doesn't list multiple concrete granular actions (e.g., specific operations like compiling circuits, simulating noise models, calibrating qubits). | 2 / 3 |
Completeness | Clearly answers both 'what' (Google quantum computing framework for noise-aware circuits, characterization experiments, low-level circuit design) and 'when' (explicit 'Use when targeting Google Quantum AI hardware, designing noise-aware circuits, or running quantum characterization experiments'). Also provides negative triggers for disambiguation. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms: 'Google Quantum AI hardware,' 'noise-aware circuits,' 'quantum characterization experiments,' 'Google hardware,' 'noise modeling,' 'low-level circuit design.' Also includes cross-references to competing tools (qiskit, pennylane, qutip) which helps with disambiguation. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with explicit disambiguation from competing quantum frameworks (qiskit for IBM, pennylane for ML/autodiff, qutip for physics simulations). Clear niche targeting Google Quantum AI hardware specifically, making conflicts very unlikely. | 3 / 3 |
Total | 11 / 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 progressive disclosure and actionable code examples. Its main weaknesses are moderate verbosity (the 'Common topics' bullet lists and best practices could be trimmed) and missing validation checkpoints in the workflow templates, particularly for hardware execution where destructive/expensive operations warrant explicit verification steps.
Suggestions
Add explicit validation steps to the hardware execution template (e.g., validate circuit against device gateset and connectivity before submission) to prevent costly errors.
Trim the 'Common topics' bullet lists under each reference section—the reference files themselves serve as the index, and these lists add token cost without actionable value.
Condense the best practices section to only non-obvious, Cirq-specific guidance; remove general advice like 'keep circuits modular' or 'document thoroughly' that Claude already knows.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary verbosity—listing 'Common topics' bullet points for each reference section is redundant given the references exist, and the best practices section restates general quantum computing wisdom Claude likely knows. However, the code examples are reasonably tight and the overall structure avoids excessive explanation. | 2 / 3 |
Actionability | The skill provides fully executable code examples for basic circuits, parameterized circuits, variational algorithms, hardware execution, and noise studies. Installation commands are concrete and copy-paste ready. The templates are complete and functional. | 3 / 3 |
Workflow Clarity | The hardware execution template lacks validation checkpoints (e.g., validate circuit against device constraints before submitting). The variational algorithm template has no error handling or convergence checking. The best practices mention 'always test on simulators first' and 'validate circuits against device constraints' but these aren't integrated into the workflow templates as explicit steps. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure with a clear quick start, then well-signaled one-level-deep references to building.md, simulation.md, transformation.md, hardware.md, noise.md, and experiments.md. Each reference section clearly describes what it contains and when to use it. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
90%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
Table of Contents
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.