# Ideate

> Generate and filter ideas into vetted options

- Skill: `synapseradio/ideate` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add synapseradio/ideate`
- Raw SKILL.md: https://api.skillmd.com/api/skills/synapseradio/ideate/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: synapseradio (https://skillmd.com/u/synapseradio)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/synapseradio/ideate

---


Follow these steps:

### 1. Seed

Decompose the opportunity, then establish:

- **The opportunity**: what could be better?
- **Hard constraints**: truly non-negotiable (timeline, budget, technical limits, values)
- **Soft preferences**: nice but not required
- **Success criteria**: how you'll recognize a good idea

Announce constraints — they become the step 4 filter.

### 2. Diverge

Generate many raw ideas without evaluating. Quantity over quality. Work through:

- **SCAMPER**: Substitute, Combine, Adapt, Modify, Put to other uses, Eliminate, Reverse
- **Lateral thinking**: break conventions, seek distant analogies
- **Inversion**: flip assumptions
- **Constraint manipulation**: add/remove constraints as catalysts
- **Wonder-driven**: follow curiosity and what-ifs

Present as a numbered list with one-line descriptions.

### 3. Cluster

Group ideas by underlying pattern — shared mechanism, same part of the problem, similar resources, or variations of each other. Reduce to a handful of named clusters, each a distinct strategic direction.

### 4. Filter

Check each cluster against the hard constraints: timeline, resources, technical viability, organizational fit. Be ruthless — don't soften constraints to save appealing ideas. Mark each: **Feasible** / **Conditional** (needs X) / **Eliminated** (fails Y).

### 5. Refine

For each surviving cluster: select the strongest variant, sketch how it works, identify the key risk, and note what makes it distinct. Enough detail to compare, not full plans.

### 6. Hand off

Present the refined options:

```
**Option [N]: [Name]**
- Approach: [one sentence]
- Strengths: [what it does well]
- Key risk: [main thing that could go wrong]
- Requires: [key dependencies]
```

Suggest next steps: clear trade-offs → weigh options; many factors → priority matrix; high stakes → premortem; stakeholder complexity → perspective shift.