tda-peer-review-panel

Runs separate specialist persona review passes on research prose and claims.

1|Updated Dec 13, 2025
One-click install
npx skills add https://github.com/ZK-Theory/TDL --skill tda-peer-review-panel-zk-theory
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tda-peer-review-panel
Source: https://github.com/ZK-Theory/TDL/tree/main/.agents/skills/tda-peer-review-panel
Command: npx skills add https://github.com/ZK-Theory/TDL --skill tda-peer-review-panel-zk-theory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Single-pass generalist reviews of research papers miss domain-specific issues in methodology, statistics, empirical plausibility, and reproducibility. This Skill structures peer review as separate specialist passes so findings are located, ranked, and converted into an actionable fix queue without premature rewriting. ## Core Features & Use Cases - Multi-Persona Review: Run Methodology, Statistical, Empirical, Reproducibility, and Prose personas as separate passes, each with its own checklist and escalation path. - Severity-Ranked Findings: Every finding cites a specific passage, is ranked blocking/major/minor/style, and cross-referenced across personas to surface converging concerns. - Use Case: When reviewing a methods section of a topological data analysis paper, select the Methodology, Statistical, and Reproducibility personas, run each separately, and produce a fix queue with required evidence per item before any rewriting begins. ## Quick Start Review the methods section of my paper using the Methodology, Statistical, and Reproducibility personas as separate passes and produce a ranked fix queue.

Frequently Asked Questions about tda-peer-review-panel

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review a research paper section with multiple specialist perspectives?

Select 2-3 personas such as Methodology, Statistical, and Reproducibility, then run each as a separate review pass rather than one combined read. Each finding must cite a specific passage and is ranked blocking, major, minor, or style before being converted into a fix queue.

What review personas are available for academic paper review?

Five personas are defined: Methodology for TDA and method validity, Statistical for inference and reporting, Empirical for social-science plausibility and data limitations, Reproducibility for regenerating results from committed artifacts, and Prose for clarity and journal register.

When should I not use a multi-persona prose review?

Do not use it for design or specification documents, which belong to adversarial design review, and do not treat it as a substitute for statistical lane audits. It targets research prose and claims, not design artifacts.

Can reviewers fix issues during the review pass?

No. Review and repair stay separate: reviewers find, cite, and rank issues, and fixing happens in a later pass. The output is a fix queue with required evidence per item, not rewritten text.

What happens when a review finding invalidates a headline result?

The review escalates: the finding is routed to computational-defect diagnosis and the lane audits before any prose fix. Blocking findings requiring methodological judgement are surfaced as user decisions with options laid out.