idea-creator-analysis

Generate and rank theory-first research ideas from papers or open problems.

2|1|Updated May 7, 2026

Search Tags:#theory #theorem #counterexample #research ideas #proof analysis #literature critique

Ready to run

npx skills add https://github.com/shenmuxing/EtaSkill --skill idea-creator-analysis

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: idea-creator-analysis
Source: https://github.com/shenmuxing/EtaSkill/tree/main/skill-examples/idea-creator-analysis
Command: npx skills add https://github.com/shenmuxing/EtaSkill --skill idea-creator-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill helps researchers turn a prompt, paper, or open problem into publishable theory-first research ideas, including theorem candidates, counterexamples, impossibility results, and proof-oriented analyses.

Core Features & Use Cases

Theory-first ideation: Produces research ideas where the main output is a formal claim, not an experiment plan.
Ranking and critique workflow: Generates a candidate set, filters it through formal sanity checks, and prepares ideas for independent critique and follow-up planning.
Use cases: Helpful when you need proof directions for a new problem, want to test whether a conjecture can become a theorem, or need a negative result or lower bound with publishable value.

Quick Start

Ask the skill to generate and rank theory-first research ideas from your problem statement or source paper, then review the top theorem candidates and proof routes it returns.

Frequently Asked Questions about idea-creator-analysis

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

FAQPage Schema

How do I generate theorem candidates from an open research problem?▼

The skill evaluates proof plausibility by applying formal sanity checks to the generated candidate set, filtering out mathematically unsound conjectures before presenting the top theorem candidates.

Can I use this to find counterexamples and lower bounds for existing conjectures?▼

Yes, you can use it to discover counterexamples and lower bounds by inputting a target theory, prompting the generation of negative results and impossibility results with publishable value.

What is theory-first research ideation and when do I need it?▼

Theory-first research ideation produces research ideas where the main output is a formal claim rather than an experiment plan, needed when proving theorems, removing assumptions, or designing diagnostic theorems.

Does generating proof-oriented research ideas require external critique?▼

External critique is optional but recommended; the core workflow generates and filters candidate sets through formal sanity checks, then prepares the ideas for independent critique and follow-up planning.

What are the limitations of automated literature critique for assumption removal?▼

Automated literature critique requires Codex-capable generation to produce rigorous candidate sets, meaning it cannot function without advanced generation capabilities and may still need independent human verification for complex proofs.