research-refine

Convert vague research directions into concrete problem anchors and method plans.

2|Updated Mar 29, 2026

Search Tags:#planning #refinement #research #proposal #foundation-model #GPT-5.4

Ready to run

npx skills add https://github.com/satsuki-64/MiniAgentWorkflow --skill research-refine-satsuki-64

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: research-refine
Source: https://github.com/satsuki-64/MiniAgentWorkflow/tree/main/.skills/research-refine
Command: npx skills add https://github.com/satsuki-64/MiniAgentWorkflow --skill research-refine-satsuki-64

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Turn vague research directions into a problem-anchored, elegant, frontier-aware, implementation-oriented method plan via iterative GPT-5.4 review.

Core Features & Use Cases

Anchor-based problem framing to anchor research direction in a concrete bottleneck.
Frontier-aware method refinement that favors minimal, high-signal interventions over bloated plans.
Iterative, critique-driven proposal generation that culminates in an actionable plan suitable for top venues.
Round-by-round revision workflow with built-in state persistence and review checkpoints.

Quick Start

Enter a vague research direction and let the pipeline generate a problem anchor and a concrete method proposal.

Frequently Asked Questions about research-refine

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

FAQPage Schema

How do I turn a vague research idea into a concrete method plan?▼

A vague research idea is converted into a concrete method plan by applying an anchor-based problem framing approach to identify a specific bottleneck, followed by iterative critique-driven refinement to generate an actionable proposal.

What is anchor-based problem framing in research proposal generation?▼

Anchor-based problem framing grounds a broad research direction in a specific, concrete bottleneck, ensuring the resulting proposal shifts from a general concept to a focused, high-signal intervention rather than a bloated plan.

How do I refine a research proposal for frontier-aware publication?▼

Refining a research proposal for frontier-aware publication involves applying iterative GPT-5.4 reviews to derive minimal, elegant solutions, ensuring the method plan meets the standards required for top academic venues.

Does this research planning workflow require a predefined validation sketch?▼

Yes, the workflow requires a lightweight validation sketch alongside a clear problem anchor and a minimal plan to effectively guide the iterative revision process and review checkpoints.

What is the best way to minimize bloated plans in foundation model research?▼

The best way to minimize bloated plans is through frontier-aware method refinement, which favors minimal, high-signal interventions over complex architectures by applying iterative, critique-driven proposal generation.

Can I use this iterative revision workflow without state persistence?▼

No, the iterative revision workflow is designed with built-in state persistence and review checkpoints to track round-by-round critiques, ensuring continuous refinement of the research proposal.