Idea Lab — Automated Idea Generation & Evaluation
Generate multiple diverse ideas, evaluate the best ones with 6-evaluator scoring, rank and surface the winners.
How It Works
/idea-lab "FinTech hackathon, 48h, team of 2, must use AI"
│
▼
Phase 0: Bootstrap ─── config.json + context setup
│
▼
Phase 1: Generate ─── N diverse ideas via forced divergence axes
│
▼
Phase 2: Pre-screen ─── quick 1-agent score per idea, filter weak ones
│
▼
══ PRE-SCREEN DASHBOARD ══ see all ideas ranked by estimate
│
▼
Phase 3: Deep Eval ─── top-K ideas → /architect (6 evaluators each)
│
▼
══ FINAL DASHBOARD ══ comparative matrix, top recommendation
│
▼
[expand N] — full evaluation details
[build N] — launch /founder-mode with the winner
[refine] — adjust context, generate more
[stop] — save and exit
Phases
| Phase |
What It Does |
Agent |
| 0 - Bootstrap |
Config generation, context capture |
bootstrapper.md |
| 1 - Generate |
N ideas with forced diversity (SCAMPER, cross-domain, axis rotation) |
generator.md |
| 2 - Pre-screen |
Quick 1-agent score per idea, filters below threshold |
prescreener.md |
| 3 - Deep Eval |
Top-K ideas evaluated via /architect (6 parallel evaluators each) |
deep-evaluator.md |
| 4 - Ranking |
Comparative matrix, scenario recommendations, top pick |
ranker.md |
Diversity Engine
Ideas are forced to diverge across configurable axes:
- B2B SaaS, B2C consumer, open-source tool, educational, mobile-first
- SCAMPER method applied per idea
- Cross-domain analogies forced for at least 2 ideas
- Anti-obvious filter removes generic suggestions
Config
Generated at .idea-lab/config.json during bootstrap:
ideaCount: how many to generate (default 5, max 20)
topK: how many advance to deep evaluation (default 3)
minScore: minimum pre-screen score to advance (default 5.0)
diversityAxes: categories for forced divergence
dimensionFilters: e.g., {"Commercial": 7} — only show ideas with Commercial >= 7
Composition
/idea-lab → best idea selected → /founder-mode → built product with GTM kit
From "I don't have an idea" to "launched product" in one pipeline.
Dependencies
- Required:
architect skill (for Phase 3 deep evaluation)
- Optional:
founder-mode (to build the winning idea)
Usage
/idea-lab "I need an idea for a FinTech hackathon, 48 hours, team of 2"
/idea-lab "SaaS product for remote teams, must be profitable in 6 months"
/idea-lab "Open source developer tool that solves a real pain point"
/idea-lab resume
Crash Recovery
State persisted to .idea-lab/state.json. If your session crashes mid-evaluation, run /idea-lab resume to continue from the next unevaluated idea.
1---2name: idea-lab3description: "Automated idea laboratory — generates multiple diverse ideas from a context, pre-screens them, evaluates the best with multi-agent scoring via /architect, and ranks the winners. Configurable idea count, diversity axes, score thresholds, and dimension filters. Composes with /founder-mode to build the winning idea. Use when: brainstorm ideas, generate ideas, hackathon ideas, idea lab, idea factory, compare ideas, evaluate multiple ideas, which idea should I build, batch evaluation, ideation."4license: MIT5metadata:6 source: https://skillmd.com/skills/juanmarchetto/idea-lab7 version: 1.0.08 category: evaluation9 tags: [ideation, brainstorming, evaluation, multi-agent, ranking, hackathon, divergent-thinking]10---1112# Idea Lab — Automated Idea Generation & Evaluation1314Generate multiple diverse ideas, evaluate the best ones with 6-evaluator scoring, rank and surface the winners.1516## How It Works1718```19/idea-lab "FinTech hackathon, 48h, team of 2, must use AI"20 │21 ▼22 Phase 0: Bootstrap ─── config.json + context setup23 │24 ▼25 Phase 1: Generate ─── N diverse ideas via forced divergence axes26 │27 ▼28 Phase 2: Pre-screen ─── quick 1-agent score per idea, filter weak ones29 │30 ▼31 ══ PRE-SCREEN DASHBOARD ══ see all ideas ranked by estimate32 │33 ▼34 Phase 3: Deep Eval ─── top-K ideas → /architect (6 evaluators each)35 │36 ▼37 ══ FINAL DASHBOARD ══ comparative matrix, top recommendation38 │39 ▼40 [expand N] — full evaluation details41 [build N] — launch /founder-mode with the winner42 [refine] — adjust context, generate more43 [stop] — save and exit44```4546## Phases4748| Phase | What It Does | Agent |49|-------|-------------|-------|50| 0 - Bootstrap | Config generation, context capture | `bootstrapper.md` |51| 1 - Generate | N ideas with forced diversity (SCAMPER, cross-domain, axis rotation) | `generator.md` |52| 2 - Pre-screen | Quick 1-agent score per idea, filters below threshold | `prescreener.md` |53| 3 - Deep Eval | Top-K ideas evaluated via `/architect` (6 parallel evaluators each) | `deep-evaluator.md` |54| 4 - Ranking | Comparative matrix, scenario recommendations, top pick | `ranker.md` |5556## Diversity Engine5758Ideas are forced to diverge across configurable axes:59- B2B SaaS, B2C consumer, open-source tool, educational, mobile-first60- SCAMPER method applied per idea61- Cross-domain analogies forced for at least 2 ideas62- Anti-obvious filter removes generic suggestions6364## Config6566Generated at `.idea-lab/config.json` during bootstrap:6768- `ideaCount`: how many to generate (default 5, max 20)69- `topK`: how many advance to deep evaluation (default 3)70- `minScore`: minimum pre-screen score to advance (default 5.0)71- `diversityAxes`: categories for forced divergence72- `dimensionFilters`: e.g., `{"Commercial": 7}` — only show ideas with Commercial >= 77374## Composition7576```77/idea-lab → best idea selected → /founder-mode → built product with GTM kit78```7980From "I don't have an idea" to "launched product" in one pipeline.8182## Dependencies8384- **Required:** `architect` skill (for Phase 3 deep evaluation)85- **Optional:** `founder-mode` (to build the winning idea)8687## Usage8889```90/idea-lab "I need an idea for a FinTech hackathon, 48 hours, team of 2"91/idea-lab "SaaS product for remote teams, must be profitable in 6 months"92/idea-lab "Open source developer tool that solves a real pain point"93/idea-lab resume94```9596## Crash Recovery9798State persisted to `.idea-lab/state.json`. If your session crashes mid-evaluation, run `/idea-lab resume` to continue from the next unevaluated idea.