skills.rest
shenmuxing
shenmuxingCOMMUNITY
Explore all skills by this author →
GitHub

idea-discovery-analysis

Orchestrate a theory-first research idea analysis pipeline with novelty screening and DeepSeek review.

2|1|Updated May 7, 2026
Search Tags:#idea generation#literature review#deepseek#research planning#theory#novelty check
One-click install
npx skills add https://github.com/shenmuxing/EtaSkill --skill idea-discovery-analysis
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: idea-discovery-analysis
Source: https://github.com/shenmuxing/EtaSkill/tree/main/skill-examples/idea-discovery-analysis
Command: npx skills add https://github.com/shenmuxing/EtaSkill --skill idea-discovery-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps you move from an open-ended theory, proof, or conceptual research question to a structured analysis pipeline that produces defensible candidate ideas, checks novelty, and turns the strongest option into an actionable plan.

Core Features & Use Cases

  • Literature Context Framing: Builds a compact source brief from the user’s prompt or a few anchor papers to ground idea generation in the right problem setting.
  • Theory-First Idea Generation: Coordinates Codex-based subagent brainstorming for proof-oriented, boundary-aware candidates rather than experiment-first hypotheses.
  • Novelty and Sanity Checks: Screens candidates for formal coherence, duplicate ideas, hidden equivalences, and prior-work overlap before committing to a direction.
  • Independent Critique and Planning: Uses DeepSeek review for adversarial evaluation and then converts the surviving idea into a concrete analysis plan with proof obligations and failure modes.

Quick Start

Use the idea-discovery-analysis skill to analyze my theory question, generate ranked candidate ideas, check novelty, and produce a final analysis plan.

Frequently Asked Questions about idea-discovery-analysis

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

FAQPage Schema
How do I generate research ideas from a theory question instead of experiments?▼

Novelty checking for mathematical research involves screening candidate ideas for formal coherence, hidden equivalences, and prior-work overlap. An adversarial critique pipeline applies independent review to validate mathematical structure and reuse safety before committing to a research direction.

What is the best way to structure a literature review for research idea generation?▼

Yes, you can convert a surviving theoretical candidate into a concrete analysis plan. After independent critique validates the idea, the pipeline transforms it into an actionable plan detailing specific proof obligations and identifying potential failure modes for the research direction.

Does this research planning pipeline require DeepSeek review and Codex subagents?▼

Yes, this research planning pipeline requires both Codex subagents and DeepSeek review. Codex subagents handle theory-first idea generation, while DeepSeek review provides independent adversarial evaluation to validate mathematical coherence and ensure reuse safety across the analysis pipeline.

What are the limitations of using automated novelty screening for theoretical proofs?▼

A limitation of automated novelty screening for theoretical proofs is its focus on proof-oriented, boundary-aware tasks rather than experiment-first validation. It screens for formal coherence and duplicate ideas but requires independent critique to fully validate mathematical structure and identify failure modes.

Related Skills

Explore More Education & Research Skills→
Me1eMe1e

tacigent-stage-solution

Identifies Tacigent Stage 3 solution direction and hands off to tacigent-stage-design.

Community
Advanced
Unified-Field-Theory-ResearchUnified-Field-Theory-Research

gpd-attack-obligation

Generate a focused attack plan for open proof obligations in Rust.

Official
Advanced
shenmuxingshenmuxing
2

deepseek-agent

Delegate critique, revision, and manuscript-writing tasks to a DeepSeek-backed OpenCode agent.

Community
Advanced
richardthe3rdrichardthe3rd
2

proof-and-analysis-toolkit

Validate code claims with empirical analysis recipes for the Cambridge Beer Festival Flutter app.

Community
Advanced