ideation-simple
Skill6 starsChanged 8 months agoMITView on GitHub
Explore a problem space and understand users - the first step in the ideation sequence
In this repository13 other files it gives its agents.
- ai-governanceSkill
- commit-phaseSkill
- comprehensive-testSkill
- diagnoseSkill
- execute-phaseSkill
- harmonizeSkill
- ideation-complexSkill
- ideation-researchSkill
- ideation-synthesizeSkill
- kickoffSkill
- plan-phaseSkill
- tech-stackSkill
- verifySkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Check if ideation.md exists in the project root.
- Distinctive: line 40
Ask: "Interesting! Before we dive into the solution, help me understand—what problem does this solve?
- Distinctive: line 72
Don't ask all of these as a list.
- Distinctive: line 76
Before writing ideation.md, confirm you have:
- Distinctive: line 81
If any of these are missing, ask one more targeted question to fill the gap before proceeding.
- Very specific: line 85
After the conversation has enough depth, create or update ideation.md in the project root with sections 1-3:
- Distinctive: line 144
Highlight key insights: Call out the most important findings from the landscape scan
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Ideation Simple
Transform a hunch into a well-understood problem with clear user personas. This is the first skill in the ideation sequence.
Prerequisites
- Web search capability: This skill uses web search to validate ideas against market reality. Ensure you have Brave Search, WebSearch, or equivalent search tools available.
- Write access: Creates
ideation.mdin the project root.
First Action: Check for Existing Ideation
Check if ideation.md exists in the project root.
If it exists, ask the user:
- "I found an existing ideation document. Would you like to:"
- "Continue where we left off"
- "Start fresh (this will replace the existing document)"
If starting fresh or no file exists, proceed with intake.
Entry Point Detection
The user may come with different starting points. Detect which type:
- Problem observed: They describe a pain point or frustration
- Example: "I'm frustrated with how I track my time at work"
- Proceed directly to problem exploration
- Solution idea: They describe something they want to build
- Example: "I want to build a timesheet app"
- IMPORTANT: Backtrack to the underlying problem first
- Ask: "Interesting! Before we dive into the solution, help me understand—what problem does this solve? What's frustrating about the current state?"
- Market opportunity: They describe a gap or trend
- Example: "Remote work teams need better tools"
- Explore who specifically has this need and what pain they experience
Phase 1: Intake & Problem Understanding
Opening question (adapt based on $ARGUMENTS if provided):
- "Tell me about what you're thinking about building—what's the spark?"
After the user shares their initial idea, identify the entry type and ensure we have a problem statement, not just a solution. Ask adaptively — if the user is forthcoming, fewer questions are needed. Dig deeper only when the picture is unclear. Goal: understand what job someone is trying to get done, how they solve it today, and what's frustrating about that.
RESEARCH INTERLUDE: Landscape Scan
Once you have a clear sense of the basic problem:
- Announce: "Let me do a quick landscape scan to see what exists in this space..."
- Search for:
- "[problem domain] solutions"
- "competitors for [problem description]"
- Synthesize findings: "Here's what I found: [summary]. This gives us context for where your idea fits."
- Note whether this validates the opportunity or reveals a crowded market.
Phase 2: User Understanding (BJ Fogg)
Understand who experiences this problem most and what would drive them to act. Ask adaptively — you're trying to understand:
- Who experiences this problem most acutely, and whether there are meaningfully different user types
- Motivation: What's the promise of a better solution? Why would they want it?
- Ability barriers: What friction might stop them from switching even if they're frustrated?
- Potential prompts: What event or moment might trigger them to seek a solution?
Don't ask all of these as a list. Surface them through natural conversation, following where the user leads.
Stopping Criteria
Before writing ideation.md, confirm you have:
- A clear problem statement (not just a solution idea)
- At least one user segment with enough context to describe them
- A rough sense of current alternatives
If any of these are missing, ask one more targeted question to fill the gap before proceeding.
Output: Create/Update ideation.md
After the conversation has enough depth, create or update ideation.md in the project root with sections 1-3:
# Ideation: [Project Name - derive from conversation]
## Meta
- Status: Simple
- Last Updated: [today's date]
- Sessions: 1
## 1. Origin
- **Entry Type**: [Problem | Solution | Opportunity]
- **Initial Statement**: [verbatim what user first said]
- **Refined Problem Statement**: [clear problem statement after exploration]
## 2. Jobs to be Done
- **Primary Job**: [the main job users are hiring a product to do]
- **Related Jobs**: [adjacent jobs that might be addressed]
- **Current Alternatives**: [how users solve this today - list with notes]
- **Switching Triggers**: [what would make them try something new]
## 3. User Segments
### Segment 1: [Name based on role/context]
- **Context**: [when/where they experience the problem]
- **Motivation**: [why they'd want a solution - the promise]
- **Ability Barriers**: [what might stop them - friction points]
- **Potential Prompts**: [what triggers action - moments of need]
### Segment 2: [If identified]
- **Context**: ...
- **Motivation**: ...
- **Ability Barriers**: ...
- **Potential Prompts**: ...
## 4. Market Landscape
[To be completed by /ideation-research]
## 5. Assumptions Log
[To be completed by /ideation-research]
## 6. Solution Hypotheses
[To be completed by /ideation-synthesize]
## 7. Open Questions for /kickoff
[To be completed by /ideation-synthesize]
## 8. Research Log
| Date | Topic | Source | Key Findings |
|------|-------|--------|--------------|
| [date] | Landscape scan | [sources] | [findings] |Completion
After generating ideation.md:
- Present summary: "Here's what we've discovered about the problem and users: [brief summary]"
- Highlight key insights: Call out the most important findings from the landscape scan
- Next step: "The simple phase is complete. When you're ready, run
/ideation-researchto dive deep into the competitive landscape and map out our assumptions."
When to Use
- Starting ideation on a new idea
- User says "I have an idea", "let's ideate", "help me think through this"
- User describes a problem, solution idea, or market opportunity
When NOT to Use
- Ideation simple already complete (Status ≥ "Simple") - use /ideation-research
- Already have formal requirements - use /kickoff directly
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

