/claim #16

Summary

• Add a self-contained AI peer-review evidence binder for the AI-Powered Research Assistant Suite.
• Validate AI-generated review comments before they are released to authors by checking concrete evidence pointers, severity calibration, unsupported accusations, blinded-review privacy leaks, and actionable remediation tasks.
• Generate reviewer-ready JSON, Markdown, SVG, and MP4 demo artifacts from synthetic data only.

Demo artifacts

• peer-review-evidence-binder/reports/demo.mp4
• peer-review-evidence-binder/reports/peer-review-evidence-report.svg
• peer-review-evidence-binder/reports/peer-review-evidence-report.md
• peer-review-evidence-binder/reports/peer-review-evidence-report.json

Validation

node peer-review-evidence-binder/test.js
node peer-review-evidence-binder/demo.js
node peer-review-evidence-binder/render-video.js
node --check peer-review-evidence-binder/index.js
node --check peer-review-evidence-binder/sample-data.js
node --check peer-review-evidence-binder/test.js
node --check peer-review-evidence-binder/demo.js
node --check peer-review-evidence-binder/render-video.js
git diff --check
ffprobe -v error -select_streams v:0 -show_entries stream=pix_fmt,width,height -show_entries format=duration,size -of default=nw=1 peer-review-evidence-binder/reports/demo.mp4

All commands passed locally. The generated MP4 reports width=960, height=540, pix_fmt=yuv420p, duration=4.000000, and size=17075.

Scope distinction

This focuses on validating the AI review output itself before author release: evidence binding, severity calibration, unsupported accusation blocking, blinded-review privacy leakage, and actionable remediation. It avoids the existing broad assistant, evidence/protocol trace, statistics review, research-gap planner, rebuttal pack, ethics/data checker, citation-context, reporting-guideline, benchmark-leakage, figure/table consistency, analysis-variable provenance, external-validity transfer, structured-abstract, omics-specific, and manuscript claim/evidence alignment slices.