skills.rest
chenghaoYang
chenghaoYangCOMMUNITY
Explore all skills by this author →
GitHub

auto-review-loop-minimax

Run iterative manuscript review loops using MiniMax with state persistence.

2|Updated Mar 19, 2026
Search Tags:#automation#workflow#research#ai-assistant#minimax#external-review
One-click install
npx skills add https://github.com/chenghaoYang/auto-coder-trainer --skill auto-review-loop-minimax-chenghaoyang
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: auto-review-loop-minimax
Source: https://github.com/chenghaoYang/auto-coder-trainer/tree/main/aris/skills/auto-review-loop-minimax
Command: npx skills add https://github.com/chenghaoYang/auto-coder-trainer --skill auto-review-loop-minimax-chenghaoyang

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomous automated external review loop for research work, leveraging MiniMax to provide structured, multi-round feedback and revision guidance to improve manuscripts without relying on Codex MCP.

Core Features & Use Cases

  • End-to-end iterative review loop: review → implement fixes → re-review, capped by MAX_ROUNDS
  • Flexible reviewer: MCP tool when available or curl fallback to MiniMax API
  • State persistence: saves REVIEW_STATE.json and AUTO_REVIEW.md to document progress
  • Logs and context: aggregates prior rounds and summary into context for reproducibility
  • Use Case: researchers preparing conference submissions can iteratively improve manuscripts based on external feedback

Quick Start

Trigger the autonomous minimax review loop by running the pipeline for your project.

Frequently Asked Questions about auto-review-loop-minimax

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

FAQPage Schema
How do I automate iterative external review for research manuscripts?▼

You can automate iterative external review by running an autonomous loop that uses MiniMax to assess manuscripts, extract feedback, and apply fixes across multiple rounds. This loop continues until reaching a capped maximum number of review iterations.

Does the automated manuscript review loop require an MCP tool to function?▼

No, the automated manuscript review loop does not strictly require an MCP tool. It uses an MCP tool when available, but automatically falls back to curl requests to the MiniMax API to ensure the review process still runs.

How does state persistence work across multiple rounds of manuscript feedback?▼

State persistence works by saving progress to REVIEW_STATE.json and logging context in AUTO_REVIEW.md. This documents each round, aggregates prior summaries, and ensures reproducibility throughout the iterative improvement cycle.

What is the best way to manage multi-round feedback for conference submissions?▼

The best way to manage multi-round feedback is using a structured review loop that processes artifacts, extracts external feedback, and tracks state. It caps iterations at a defined maximum limit to prevent endless cycles while improving manuscript quality.

Can I use this automated review process without relying on Codex MCP?▼

Yes, you can use this automated review process without relying on Codex MCP. It specifically leverages MiniMax to provide structured feedback and revision guidance, utilizing an MCP tool or curl fallback instead of Codex.

Related Skills

Explore More Education & Research Skills→
dz306271098dz306271098

auto-review-loop-minimax

Orchestrate multi-round external review of ML research proposals using MiniMax.

Community
Advanced
SimonZeng7108SimonZeng7108
17

recomb

Format RECOMB manuscripts with Springer LNCS guidelines and splncs04 BibTeX style.

Community
Intermediate
shishishishi
3

codex-review

Automate iterative code reviews with Codex and Claude Code patches.

Community
Advanced
czh-ee-2023czh-ee-2023

auto-review-loop-minimax

Automate multi-round research review loops using the MiniMax API.

Community
Advanced