Reproducibility auditor
Skill maximussthegreat/ml-researcher-os/skills/reproducibility-auditor
Agent skills and workflows for reproducible ML research.
npx -y skills add maximussthegreat/ml-researcher-os --skill reproducibility-auditorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when auditing an ML repository, experiment folder, model release, or paper reproduction for missing files, hidden assumptions, and rerun readiness.
SKILL.md
1.1 KB, as published. Nobody here has run it
Reproducibility Auditor
Use this skill before publishing results, releasing a model, or asking others to reproduce a run.
Audit checklist
Check for:
- environment file
- dependency versions
- dataset source and license
- split creation code
- preprocessing code
- seed policy
- training config
- evaluation config
- logs
- checkpoints
- result table
- model card or report
- known limitations
Required output
| Item | Status | Evidence | Required fix |
|---|
Status must be one of:
presentmissingunclearnot_applicable
Reproducibility grade
End with:
Grade: A/B/C/D/F
Can a stranger rerun this? yes/no/partial
Smallest next fix: ...
Rules
- Prefer one small fix that unlocks reruns over a large refactor.
- Do not accept screenshots as a substitute for logs.
- Do not accept "works on my machine" as evidence.