| name | dual-column-self-check |
| description | Run one of the ML/CS reproducibility checklists (ML Reproducibility Checklist, REFORMS, NeurIPS Paper Checklist, Model Cards, Datasheets for Datasets) against a paper as a reader-side audit, producing a category (Yes/No/NA) plus free-text reason per item. Use this whenever the user wants a reproducibility/completeness self-check run on an ML or CS paper — invoke this directly, it has no study-design gate in this package since these checklists are engineering self-audits, not clinical-study tools. |
| version | 1.0.0 |
| category | paper-reading |
| type | sop |
| execution | subagent |
| prompt | ./prompt.md |
| input | source_path (string), meta_path (string), item_set (string — name of the specific checklist) |
| reads | full paper — a completeness self-audit asks whether each item appears anywhere |
| output | checklist_result (list of {item, category, reason}) |
| dependencies | {"sops":["spawn-agent"]} |
Dual Column Self-Check
Category (Yes/No/NA) + free-text reason per item, across 5 ML/CS reproducibility checklists. Originally author-facing self-certification tools, reversed here for reader-side auditing — each item's framing must be flipped to a question before being answered.
Execution
Subagent — spawned via spawn-agent skill.
No upstream gate (intentional, not a gap)
Unlike quality-appraisal-checklist/reporting-standard-checklist, this SOP has no in-edge from study-design-tool-gate in the graph — its 5 checklists are ML/CS engineering self-audits, not tied to a clinical study design, so no study-design dispatch was ever drawn to it (spec §5's flagged note). Do not add a gate dependency here without revisiting that decision explicitly.
Available SOPs
| SOP | When to use |
|---|
| spawn-agent | Spawn a customized CC subagent with full MCP tool access. |