auto-review-loop-minimax

Coordinate multi-round external reviews via the MiniMax API with persistent state.

Updated Apr 18, 2026
One-click install
npx skills add https://github.com/THUFanZd/Sewed_pipeline --skill auto-review-loop-minimax-thufanzd
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-review-loop-minimax
Source: https://github.com/THUFanZd/Sewed_pipeline/tree/main/.agents/skills/auto-review-loop-minimax
Command: npx skills add https://github.com/THUFanZd/Sewed_pipeline --skill auto-review-loop-minimax-thufanzd

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomously coordinate multi-round external reviews using the MiniMax API to drive iterative improvements, reducing manual back-and-forth and surfacing actionable convergence signals.

Core Features & Use Cases

  • Multi-round review loop: Review → implement fixes → re-review until acceptance or max rounds.
  • MiniMax integration: Leverages the MiniMax API for external, independent assessment.
  • Stateful execution: Persists progress in REVIEW_STATE.json and logs rounds in AUTO_REVIEW.md for traceability.
  • Flexible execution modes: Supports MCP tool when available or curl-based fallback to Minimax API.
  • Ideal use cases: Manuscript improvement, research proposal refinement, and quality assurance for research outputs.

Quick Start

Run the auto-review-loop-minimax skill to initiate an autonomous MiniMax-based external review cycle.

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 multi-round external reviews for research manuscripts?

Automating multi-round external reviews is achieved by coordinating an autonomous loop where the MiniMax API reviews research outputs, applies fixes, and re-reviews until acceptance. Progress is persisted in REVIEW_STATE.json for traceability.

What is an autonomous review loop and how does it improve research proposals?

An autonomous review loop uses the MiniMax API to provide independent, iterative assessments of research proposals. It drives improvements by cycling through review and fix stages until convergence signals are met, reducing manual back-and-forth feedback.

Can I use curl to call the MiniMax API for automated manuscript feedback?

Yes, you can use curl to call the MiniMax API for automated manuscript feedback. The workflow supports flexible execution modes, utilizing the MCP tool when available or falling back to curl-based API requests.

Do I need a MiniMax API key to run iterative research output reviews?

Yes, you need MiniMax API access to run iterative research output reviews. The workflow leverages this external API for independent assessment and requires persistent state via REVIEW_STATE.json to track progress.

How do I track the progress of an automated MiniMax review cycle?

You track the progress of an automated MiniMax review cycle using stateful execution. The workflow persists progress in REVIEW_STATE.json and logs detailed round histories in AUTO_REVIEW.md for full traceability.

When should I not use an autonomous external review loop for research outputs?

You should not use an autonomous external review loop when your research outputs require subjective human judgment beyond API assessment, or when you lack persistent state storage for REVIEW_STATE.json and MiniMax API access.