Idea Generation Skill Summary
SkillDev toolsA skill for dev tools by lamm-mit.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Idea Generation Skill Summary skill
What this skill tells your AI
The instructions your AI receives, as published by lamm-mit/scienceclaw in skills/idea-generation/SKILL.md and read by ahel’s review.
This workflow generates 5 research ideas grounded in literature, following this process:
Key Steps
- Check Workspace – Audit existing papers in
$W/papers/ - Plan Strategy – Decide whether to use current collection or search for more papers
- Acquire Resources – Either delegate to
/research-collect(100+ papers) or run quick searches - Analyze Literature – Extract contributions, methods, limitations, and gaps from papers
- Generate Ideas – Create 5 distinct concepts using different strategies (combination, simplification, generalization, constraint relaxation, architecture innovation)
- Score & Enhance – Rate ideas by novelty/feasibility/impact; detail the top candidate with math and implementation roadmap
- Code Mapping – Link concepts to reference implementations
- Summarize – Document all findings and next steps
Core Principle
"Ideas MUST be grounded in actual papers, not generated from model knowledge." Each idea requires citations (≥2 papers by arXiv ID).
Output Location
Results save to $W/ideas/ as individual markdown files, plus consolidated summary.
Integration
Works with /research-collect (upstream) and /research-pipeline (downstream for implementation).
Signals
- GitHub stars
- 242
- Forks
- 42
- Last commit
- Aug 2026
Advanced
- Catalog kind
- skill
- Gateway key
idea-generation-lamm-mit- Source
- github.com/lamm-mit/scienceclaw