qiskit
IBM quantum computing framework. Use when targeting IBM Quantum hardware, working with Qiskit Runtime for production workloads, or needing IBM optimization tools. Best for IBM hardware execution, quantum error mitigation, and enterprise quantum computing. For Google hardware use cirq; for gradient-based quantum ML use pennylane; for open quantum system simulations use qutip.
87
81%
Does it follow best practices?
Impact
99%
1.39xAverage 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 with excellent completeness and distinctiveness. The explicit 'Use when' clause with specific triggers and the cross-references to alternative tools for different use cases make it highly effective for skill selection. The main weakness is that the specific capabilities could be more granular—listing concrete actions like 'build quantum circuits, transpile for hardware backends, submit jobs to IBM Quantum' rather than broad capability areas.
Suggestions
Replace broad capability areas like 'IBM hardware execution' and 'enterprise quantum computing' with more concrete actions such as 'build quantum circuits, transpile for IBM backends, submit jobs to IBM Quantum, apply error mitigation techniques.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (IBM quantum computing) and some actions like 'IBM hardware execution, quantum error mitigation, and enterprise quantum computing,' but these are more like capability areas than concrete specific actions. It doesn't list granular operations like 'build circuits, transpile for hardware, submit jobs, analyze results.' | 2 / 3 |
Completeness | Clearly answers both 'what' (IBM quantum computing framework for hardware execution, error mitigation, optimization tools) and 'when' (explicit 'Use when targeting IBM Quantum hardware, working with Qiskit Runtime for production workloads, or needing IBM optimization tools'). Also includes negative routing guidance for alternative tools. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms: 'IBM Quantum', 'Qiskit', 'Qiskit Runtime', 'IBM hardware', 'quantum error mitigation', 'enterprise quantum computing', plus cross-references to competing tools (cirq, pennylane, qutip) which helps with routing. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with explicit differentiation from competing quantum frameworks (cirq for Google, pennylane for gradient-based ML, qutip for open quantum systems). This cross-referencing makes it very unlikely to conflict with other quantum computing skills. | 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 some unnecessary marketing/promotional content in the overview that wastes tokens, and a lack of explicit validation checkpoints in workflows (e.g., checking job status, handling transpilation errors, verifying results). The decision guide and reference structure are particularly effective for navigation.
Suggestions
Remove marketing statistics ('13M+ downloads', '83x faster transpilation', '29% fewer two-qubit gates') from the overview—these don't help Claude execute tasks and waste tokens.
Add explicit validation/error-handling steps to the Common Patterns, such as checking job.status() before retrieving results and handling TranspilerError exceptions.
Trim the topic bullet lists under each reference section heading—the reference file names and one-line descriptions are sufficient for navigation without enumerating every sub-topic.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The overview includes some marketing-style content ('world's most popular', '13M+ downloads', '83x faster transpilation') that doesn't help Claude execute tasks. The topic lists under each reference section are somewhat redundant given the reference files exist. However, the code examples themselves are lean and the workflow decision guide is efficient. | 2 / 3 |
Actionability | The skill provides fully executable, copy-paste ready code examples for the most common patterns (Bell state, hardware execution, VQE). The Quick Start, Common Patterns, and Best Practices sections all contain concrete, runnable Python code with specific imports and method calls. | 3 / 3 |
Workflow Clarity | The Development Workflow section provides a clear sequence (simulators → transpile → choose primitives → choose execution mode) and the Common Patterns show progressive complexity. However, there are no explicit validation checkpoints or error recovery steps—e.g., no guidance on checking transpilation results, handling failed jobs, or verifying hardware execution succeeded before proceeding. | 2 / 3 |
Progressive Disclosure | Excellent progressive disclosure structure: a concise overview with quick start code, followed by clearly signaled one-level-deep references to 8 topic-specific files. The Workflow Decision Guide provides an effective navigation aid mapping user intent to the correct reference file. | 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
b58ad7e
- 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.