idea-creator

Generate ranked research ideas with minimal viable experiments from a research direction.

2|1|Updated Apr 19, 2026
One-click install
npx skills add https://github.com/raja21068/AutoResearch --skill idea-creator-raja21068
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: idea-creator
Source: https://github.com/raja21068/AutoResearch/tree/main/skills/aris/idea-creator
Command: npx skills add https://github.com/raja21068/AutoResearch --skill idea-creator-raja21068

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you convert a broad research direction into concrete, publishable experiment-ready ideas instead of vague brainstorming that never turns into actionable work.

Core Features & Use Cases

  • Landscape survey: scans existing local papers first, then searches recent literature to map approaches and gaps.
  • Idea generation + ranking: generates multiple hypotheses with minimal viable experiments, filters by feasibility and novelty signals, and ranks survivors.
  • Pilot validation: optionally runs short pilot experiments for top ideas to re-rank based on real empirical signal, not just text reasoning.
  • Wiki-ready outputs: produces a structured ranked idea report and can write idea pages plus edges into a research wiki when available.

Quick Start

Use idea-creator to generate ranked research ideas by providing your research direction as an argument.

Frequently Asked Questions about idea-creator

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

FAQPage Schema
How do I generate publishable research ideas from a broad direction?

Generate publishable research ideas by surveying local paper libraries, discovering recent literature, and identifying gaps. You get ranked hypotheses with minimal viable experiments ready for execution.

What is the best way to check research novelty and feasibility before running experiments?

Check research novelty by scanning existing local papers and recent literature to map approaches. You get filtered hypotheses ranked by feasibility signals and novelty, validated by optional short pilot experiments.

How do I use a local paper library to plan a multi-agent research workflow?

Plan a multi-agent research workflow by parsing your local paper library for landscape surveys and using an external LLM via Codex MCP for brainstorming. You get a structured ranked IDEA_REPORT output.

Can I run pilot experiments to validate ranked research hypotheses?

You can run short pilot experiments for top ideas by budgeting compute to gather empirical signal. You get re-ranked research ideas based on real validation results instead of text reasoning alone.

Does idea generation and experiment design work for machine learning and adjacent academic domains?

Idea generation and experiment design works for ML and adjacent academic domains by proposing testable hypotheses. You get minimal viable experiments tailored to your specific research planning requirements.