auto-paper-improvement-loop

Iteratively review and correct LaTeX papers through external LLM feedback loops.

Updated Jun 7, 2026
One-click install
npx skills add https://github.com/czh-ee-2023/zotero-aris --skill auto-paper-improvement-loop-czh-ee-2023
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-paper-improvement-loop
Source: https://github.com/czh-ee-2023/zotero-aris/tree/main/.claude/skills/auto-paper-improvement-loop
Command: npx skills add https://github.com/czh-ee-2023/zotero-aris --skill auto-paper-improvement-loop-czh-ee-2023

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mcp__codex__codex, python, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenge of improving the quality of a generated paper through iterative reviews and corrections, ensuring high-quality writing and content.

Core Features & Use Cases

  • Iterative Review and Correction: Automatically improve a paper through multiple rounds of review by a GPT-5.5 model, applying fixes, and recompiling.
  • Automated Loop: Runs an internal loop (review → fix → recompile) for up to two rounds, ensuring structural and presentation improvements.
  • Optional Style Reference: Allows users to reference a paper style guide for structural fixes only during the improvement phase.
  • Optional Edit Whitelist: Enables users to constrain which files and operations can be edited by the fix-implementation step.

Quick Start

To start the paper improvement loop, run the command: /auto-paper-improvement-loop "paper/"

Frequently Asked Questions about auto-paper-improvement-loop

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

FAQPage Schema
How do I automate LaTeX paper improvement through iterative review?

Automated paper improvement runs an iterative loop of external LLM review and correction on LaTeX source files, applying fixes and recompiling the PDF for up to two rounds to ensure structural and presentation enhancements.

Can I use a style reference to guide automated paper correction?

Yes, automated paper correction supports an optional style reference paper, allowing the GPT-5.5 model to apply structural fixes during the improvement phase to match your desired formatting and presentation standards.

Do I need Codex MCP to run the automated paper improvement loop?

Yes, the paper improvement loop requires the Codex MCP for external LLM review and uses Python for file manipulation and PDF recompilation to execute iterative corrections on your LaTeX documents.

How can I restrict which files the automated paper correction edits?

You can use an optional edit whitelist to constrain which files and operations the fix-implementation step modifies, giving you fine-grained control over the automated paper improvement process.

What is the best way to fix compiled LaTeX PDF quality issues automatically?

The best way to fix compiled LaTeX PDF quality issues is running an iterative review and correction loop, where an external GPT-5.5 model reviews the paper, applies structural fixes, and recompiles the source files.

Are there limitations to the iterative paper review and correction loop?

The iterative paper review and correction loop is limited to two automated rounds of LLM review, fixes, and recompilation, meaning highly complex or deeply flawed LaTeX documents may require additional manual intervention.