academic-paper-reviewer

Simulates a five-seat peer review panel producing review reports and editorial decisions for academic papers.

Updated Mar 23, 2026
One-click install
npx skills add https://github.com/hanasho744/codex --skill academic-paper-reviewer-hanasho744
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: academic-paper-reviewer
Source: https://github.com/hanasho744/codex/tree/main/.agents/skills/academic-paper-reviewer
Command: npx skills add https://github.com/hanasho744/codex --skill academic-paper-reviewer-hanasho744

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Authors lack access to realistic pre-submission peer review, so weaknesses in methodology, argumentation, and journal fit surface only after formal submission. This Skill simulates a complete journal review process with multiple role-separated reviewer personas, delivering structured critique and an editorial decision before the real submission. ## Core Features & Use Cases - Multi-Perspective Review Panel: Runs a five-seat panel (Journal-Fit Reviewer, Methodology, Domain, Perspective, and Devil's Advocate reviewers) with field-specific personas configured automatically from the paper's discipline. - Six Operational Modes: Supports full review, re-review (revision verification), quick assessment, methodology-focus, Socratic guided review, and calibration modes for measuring reviewer decision-error profiles. - Editorial Synthesis: Consolidates all reviewer reports into an Editorial Decision Letter and an immutable Revision Roadmap, with visible adjudication of Devil's Advocate critical findings. - Use Case: A researcher pastes a draft manuscript and receives five independent review reports plus an editorial decision and revision roadmap, then uses re-review mode after revising to verify each comment was addressed. ## Quick Start Ask the assistant to review this paper and paste the manuscript text or provide the paper file.

Frequently Asked Questions about academic-paper-reviewer

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

FAQPage Schema
How do I get a simulated peer review of my paper before submission?▼

Paste your manuscript or provide the file and ask for a paper review. The skill identifies the field, configures four reviewer personas plus a fixed Devil's Advocate, and returns five review reports with an editorial decision and revision roadmap.

What review modes does academic paper review support?▼

Six modes are available: full review, re-review for verifying revisions, quick 15-minute assessment, methodology-focus, guided Socratic review, and calibration mode that measures decision-level error rates against adjudicated papers.

Can I verify that my revisions addressed the reviewer comments?▼

Yes, re-review mode checks revisions against the original review comments using a traceability matrix. It requires the original roadmap, author adjudication sidecar, revision evidence, and both manuscript versions, then issues a new decision.

Does the simulated review modify my manuscript?▼

No, reviewers operate under a read-only constraint and never edit the submitted manuscript. All output is produced as separate documents: review reports, the editorial decision letter, and the revision roadmap.

What are the limitations of AI-simulated peer review?▼

Reviews are criterion-bound judgements with calibration status NOT_CALIBRATED, meaning no measured error profile applies to live reviews. Role separation between personas is not a claim of independent error processes, and the panel does not replace human editorial judgement.