idea-creator

Generate and validate research ideas through literature review and pilot experiments.

Updated Jul 6, 2026
One-click install
npx skills add https://github.com/caw111/2026-SoftwareCup --skill idea-creator-caw111
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: idea-creator
Source: https://github.com/caw111/2026-SoftwareCup/tree/main/.agents/skills/idea-creator
Command: npx skills add https://github.com/caw111/2026-SoftwareCup --skill idea-creator-caw111

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires codex, gpt-5.5, research-wiki, GPU, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill helps researchers systematically generate, validate, and rank publishable research ideas, reducing the time and effort spent on initial idea exploration.

Core Features & Use Cases

  • Research Idea Generation: Automatically generate concrete research ideas based on a broad research direction.
  • Novelty and Feasibility Check: Validate ideas for novelty and feasibility through multiple stages, including literature review, brainstorming, and pilot experiments.
  • Parallel Pilot Experiments: Run parallel pilot experiments to gather empirical evidence for top ideas.
  • Structured Reporting: Generate a structured report with ranked ideas, landscape summary, and pilot results.

Quick Start

Generate research ideas for the topic 'neural networks' by running the command: /idea-creator "neural networks" --pilot budget: 4h per idea, 20h total.

Frequently Asked Questions about idea-creator

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

FAQPage Schema
How do I systematically generate and validate research ideas for a new topic?

You can generate research ideas by providing a broad research direction to an automated system that performs literature reviews, brainstorming, novelty checks, and feasibility assessments to output structured, ranked ideas.

What is the best way to check the novelty and feasibility of research ideas?

Checking novelty and feasibility of research ideas requires a multi-stage validation process combining literature reviews, automated novelty checks, feasibility assessments, and parallel pilot experiments to ensure publishability.

Can I run parallel pilot experiments to test multiple research ideas at once?

Yes, you can run parallel pilot experiments to test multiple research ideas simultaneously by allocating a specific pilot budget and utilizing GPU resources to gather empirical evidence for top ranked ideas.

Do I need GPU resources to perform automated feasibility assessments for research ideas?

Yes, you need GPU resources to perform automated feasibility assessments because the validation process requires running parallel pilot experiments to gather empirical evidence for the generated research ideas.

How does a Codex agent review and validate generated research ideas?

A secondary Codex agent reviews and validates generated research ideas by acting as an automated reviewer within a multi-step workflow, performing novelty checks and feasibility assessments alongside a research wiki.

What are the limitations of using automated research idea generation for academic publishing?

Limitations of automated research idea generation include dependency on GPU resources for pilot experiments, requiring a secondary Codex agent for review, and needing a research wiki to accurately validate novelty and feasibility.