Phase 1 of 6

Run /visionaire:start

No blank page. No guessing what to say next. Describe what you're building — the right questions come to you.

The space for freeform exploration before rigor asks you to commit. Capture the excitement. Surface the assumptions. Name what you're not building.

Start my spec session — $297
$297 early access — first 100 builders  ·  Becomes $447 after
See how it works  →
One payment. Keep it forever. Runs locally. No telemetry.
01
Idea
IDEA.md
Capture the excitement before analysis kills it.
IDEA.md
BRIEF.md
MARKET-analysis.md
1# IDEA.md — Async Team Dashboard
2
3## The Core Idea
4A Slack-native view that surfaces who's blocked, who's
5available, and what decisions need input — without requiring
6anyone to narrate their status manually.
7
8## Why I Want to Pursue This
9 Teams I've led lose ~2 hrs/person/week to sync that
10 could be async. This problem is personal.
11 Success: manager opens Slack and knows team state
12 without sending a single message.
13
14## Key Assumptions ← To Validate in Product Brief
151. Slack API exposes the signals we need
162. Teams will opt into passive signal collection
17
18## What We're NOT Building
19→ Another standup bot
20→ A project management layer over Jira
21✓ Core idea crystallized
22✓ Motivation documented
23✓ Assumptions named for Product Brief to validate
24→ IDEA.md created · Product Brief phase ready
If you skip this phase
  • You will start building after a "clear enough" idea that isn't actually clear.
  • You will make decisions quickly to keep momentum, not because they are correct.
  • You will realize something fundamental is wrong only after multiple parts depend on it.

None of this is theoretical. These are the patterns that show up every time.

Most builders skip this and open a blank document. Three hours later they call it done when they're tired.

Skipping ideation doesn't just cost you a better idea — it costs you the record of why you chose what you chose. When direction needs to change three weeks in, there's no documented exploration to return to. You start the conversation over, from nothing.

The first-idea trap
You lock into the first viable approach before exploring whether a better one exists. Three weeks in, someone describes something clearly superior. You're already committed — code written, decisions made.
Motivation that disappears mid-build
Forced into analytical mode from the start, you lose touch with why you wanted to build this. Two months in, the bugs multiply. You read the spec looking for what made this worth doing. It's not there.
Assumptions that become invisible architecture
You assumed desktop-first. You assumed real-time over async. These beliefs quietly shaped your architecture before anyone questioned them. Months of work built on an unexamined belief.
Features that feel bolted on
For existing apps, skipping ideation means integration was never considered at the conceptual level — only the technical one. Users feel it. The product loses coherence.

Every specification gap catches 3–8 hours of rework. Visionaire catches an average of 12 per project. Do that math once.

From messy to crystallized — without losing what made the thinking possible.

The Idea Phase takes the contradictions, tangents, and "what if" thinking and produces a structured artifact without compressing the exploration that generated it.

Brainstorm
What if we...? Variations Contradictions Tangents Why does this matter? Excited about... What about async? Different approach? Assumptions?
IDEA.md
Core idea — crystallized
Why you care — documented
Variations explored — visible
Assumptions — named
Open questions — explicit
What you're not building — defined

How this phase is built — not what it does, but how it behaves every time.

The Idea Phase is the only phase in the system without a formal gate. These are the structural commitments that replace it.

01
You can be messy here
No wrong answers. Rambling is welcome. Tangents are where insights live. This is the one phase that explicitly doesn't require rigor.
02
Nothing locks you in
Everything here is context, not constraint. IDEA.md doesn't bind later phases — they can deviate completely. Explore widely.
03
Your excitement is documented
The personal "why" gets captured explicitly. Six months from now, when momentum flags, it's there. Not just remembered. Written down.
04
Assumptions get named first
Every idea contains hidden assumptions. This phase surfaces and tags them for validation. Nothing gets built on unexamined beliefs.
05
Every path you considered stays visible
"We chose X over Y because..." doesn't disappear. When direction hits problems, you return to documented alternatives — not scratch.
06
A conversation, not an interrogation
No fixed questions. No required sequence. The conversation follows your energy. Insights live in the tangents — so this phase doesn't cut them off.
07
Existing apps get understood
Adding to something already built? The phase can explore your codebase before helping you think through how the new idea fits. Grounded ideation, not speculation.
08
You decide when you're done
No checklist. No required outline. You signal readiness. The framework waits.
Every gap you leave in an idea becomes an assumption the LLM builds on. Silently. In a direction you never chose.

The Idea Phase is where those gaps get found — before Product Brief builds on them.

Start my spec session — $297
$297 early access — first 100 builders  ·  Becomes $447 after
One payment. Yours forever. Runs locally in Claude Code. 30-day guarantee.

Conversation to document in four steps.

Unlike every other phase, the Idea Phase uses conversational approval. The summary emerges from the conversation — not from filling a template.

01

Freeform exploration

You talk. The agent listens, asks follow-up questions, surfaces assumptions you didn't know you were making, and helps you explore variations you hadn't considered. No fixed sequence. Continues until you signal readiness.

02

Summary presented in chat

When you're ready, the agent generates a summary — not a file yet. Core idea, what excites you, variations explored, assumptions to validate, open questions, and explicit boundaries you set.

03

You respond

Approve the summary, request corrections, or keep exploring. This is the only review gate in the phase — and you control it. The loop repeats as many times as needed.

04

IDEA.md created

Once approved, the document is created. Not before. You now have a crystallized artifact that the Product Brief phase builds on — or consciously deviates from based on what it learns.

Note

No automated gate. The Idea Phase is the only phase in Visionaire without a formal quality-gate reviewer. You self-determine readiness. Every other phase — Product Brief, Market Validation, Architecture, Features — has a specialized critic that must approve output before you advance.

Where you are before you run this phase — it handles all three.

Not an idealized greenfield project. Here's how it actually works.

Scenario 01

You have a vague product idea

You know what excites you. But when someone asks you to articulate it precisely, something gets lost. The pressure to define everything before you've had time to think is killing the idea before it starts.

The Idea Phase gives you space to explore without that pressure. The conversation follows what excites you — variations get surfaced, assumptions get named, and the personal "why" gets captured alongside the product idea itself.

What you walk away with

A crystallized IDEA.md: what excites you, what variations you considered, what you're assuming, and what you've explicitly decided not to build. Ready for Product Brief to validate.

Scenario 02

You're adding a feature to an existing app

You have users. You have a codebase. Specs written without understanding what already exists produce features that feel foreign. You need ideation grounded in what's actually there.

When you mention an existing application, the Idea Phase can explore your codebase — current features, architecture, constraints — before helping you think through how the new idea connects.

What you walk away with

An IDEA.md that grounds the new idea in what exists — how it extends, complements, or transforms what you've already built. Not a feature bolted on. A coherent addition.

Scenario 03

You need to pivot after a rejection

Product Brief Review came back hard. Market Validation found 47 competitors and unclear differentiation. Starting completely over feels wasteful — you've learned things that should inform what comes next.

The Idea Phase can be re-entered with market feedback incorporated. The conversation starts with what you learned, explores pivots, and produces IDEA-v2.md — preserving the exploration history while capturing the new direction.

What you walk away with

IDEA-v2.md: why the original direction didn't work, what variations were considered in response, and which refined direction Product Brief will now validate. The history is preserved, not abandoned.

Phase 1 of six. The widest point of the funnel.

Where possibilities are expansive before they're narrowed through validation and structure. Product Brief inherits assumptions — it doesn't surface them.

Phase 1IdeaYou are here
Phase 2Product Brief
Phase 3Market Validation
Phase 4UI/UX
Phase 5Architecture
Phase 6Features

You can skip this and move straight to Product Brief.

Teams do it every day. The tradeoffs are predictable.

Without the Idea Phase

Faster start, higher downstream cost

  • Product Brief validates the first version of an idea that never had space to improve
  • Hidden assumptions get built into architecture before anyone names them
  • The motivation that sustains builders never gets documented
  • Variations explored informally disappear — pivots feel like starting over
  • Features for existing apps get designed without codebase context
  • Six months in, no record of what made this worth doing
With the Idea Phase

Slower start, clearer foundation

  • Product Brief has rich context — what you're excited about, what you're assuming, what you've already considered
  • Assumptions become explicit validation targets from day one
  • The "why" is preserved — documented, accessible, recoverable when needed
  • Exploration history stays visible — pivots build on what came before
  • Existing app context grounds new ideas in what's actually there
  • Every phase that follows operates on explored thinking, not a first draft

This phase doesn't slow Product Brief down. It changes what Product Brief is working with.

A Product Brief built on explored, assumption-surfaced thinking produces sharper output than one built on the first version of an idea that never had space to breathe. The hour you spend here isn't overhead. It's the decisions that make every phase after it faster.

Two principles that shaped every decision in this phase.

"When we fail to define something our customer wants, we fail to open a story gap. When we don't open a story gap in our customers' mind, they have no motivation to engage us, because there is no question that demands resolution."

— Donald Miller · Building a StoryBrand

Before you open a story gap for your customers, you have to open one for yourself: What excites you about this idea? What problem are you drawn to solve? These aren't analytical questions. They're motivational. The Idea Phase surfaces them before market analysis buries them.

"The best ideas emerge from a state of play, not a state of analysis. Analysis narrows. Play expands. You need both — but in the right order."

— Ed Catmull · Creativity, Inc.

The Idea Phase is built on this sequencing. Exploration before constraint. You can't expand possibilities inside an analytical frame — the frame is what makes you narrow too early. The system asks questions in conversation form precisely because conversation stays open. The document comes later. After the thinking has had room.

Every project you start without this phase is a first draft that never had space to become more.

The teams that skip it don't get bad results — they get good results on the wrong thing. That gap is what Visionaire exists to close.

The questions Visionaire asks you aren't a fixed list. They change based on what you say — the same way a good interviewer does.

Every gap you leave in an idea becomes an assumption the LLM builds on. Silently. Confidently. In a direction you never chose. This phase is where those gaps get found.

— Robert Evans

No blank page. No guessing what to say next.

Describe what you're building. The right questions come to you. The gaps get found before Product Brief inherits them. The thinking gets documented before the pressure to decide closes it down.

$297 < 2 hours of senior dev time · saves 5–10× that per project
Start my spec session — $297
$297 early access — first 100 builders  ·  Becomes $447 after
One payment. Yours forever. Runs locally in Claude Code. 30-day guarantee.
View Sample IDEA.md  ↑ Read Product Brief Phase  →