| name | jupyter-reproducibility-checker |
| description | Skill for checking and ensuring Jupyter notebook reproducibility |
| allowed-tools | ["Bash","Read","Write"] |
| metadata | {"specialization":"scientific-discovery","domain":"science","category":"Reproducibility","skill-id":"SK-SCIDISC-024"} |
| graph | {"domains":["domain:scientific-discovery"],"specializations":["specialization:scientific-research-methods"],"skillAreas":["skill-area:data-analysis","skill-area:statistical-analysis","skill-area:deep-web-research"],"workflows":["workflow:experiment-design","workflow:peer-review-cycle"],"roles":["role:research-engineer","role:computational-scientist"]} |
Jupyter Reproducibility Checker Skill
Purpose
Check and ensure reproducibility of Jupyter notebooks including cell execution order, environment dependencies, and output consistency.
Capabilities
- Verify execution order
- Check dependencies
- Test reproducibility
- Clear and rerun notebooks
- Document environments
- Generate requirements
Usage Guidelines
- Load notebook
- Check execution order
- Identify dependencies
- Test fresh execution
- Document environment
- Generate reports
Process Integration
Works within scientific discovery workflows for:
- Reproducibility audits
- Notebook cleanup
- Environment documentation
- Quality assurance
Configuration
- Check criteria
- Execution settings
- Environment capture
- Report formatting
Output Artifacts
- Reproducibility reports
- Dependency lists
- Environment files
- Cleaned notebooks