Decision exploration through pattern recognition and progressive discovery
Current Status: Stage 1 (Explore) complete - v4.0 shipped
/explore [your decision]
Example:
/explore Should I take this $6K/month client work or focus on my product?
Outcomist will:
- Recognize the pattern - Reveal what's really being decided (30 sec)
- Ask targeted questions - Build context one question at a time (5-10 min)
- Deliver clear recommendation - Specific to YOUR situation with next actions
A decision exploration tool that helps you understand what you're REALLY deciding.
People frame decisions wrong:
- "Should I redesign my office?" → Really about creating personal sanctuary
- "Should I take stable money or bet on my product?" → Really about scarcity mindset despite strong fundamentals
- "Should I build for 10 customers or 2?" → Really about product identity crisis
Outcomist reveals the real question through pattern recognition and adaptive discovery.
Outcomist is Stage 1 of a three-stage decision journey:
Understand the real problem through pattern recognition and discovery
- Uncover: When you don't know the core problem
- Inform: When context needs synthesis
- Challenge: When you think you know but you're wrong
Output: Understanding + clear recommendation
Align on what to build before building it
- Show wireframes, designs, specs, prototypes
- Iterate on OUTPUT before execution
- Prevents "building the wrong thing efficiently"
Output: Visual representation you sign off on
Execute and track implementation
- Actually build the thing
- Handle blockers
- Measure outcomes
Output: Working solution
- Product Vision - What we're building & why
- Three-Stage Journey - explore → design → build (coming soon)
- User Personas - Who this is for (coming soon)
- Pattern Library - 6 patterns captured from testing
- Question Framework - Question quality heuristics
- Progressive Disclosure - Tier 1-4 architecture
- Test Scenarios - 20 scenarios for validation
- MUST_HAVES - Active implementation roadmap
- Test Logs - Historical test documentation
- Changelogs - Version history
Shipped: v4.0
- ✅ Pattern recognition working
- ✅ Discovery questions validated
- ✅ 5 test scenarios completed
- ✅ 6 patterns captured
- ✅ User validation consistently positive
Next: Complete validation
- ⏳ 5 more test scenarios (50% complete)
- ⏳ Pattern library (60% complete - target 10+)
- ⏳ Question framework implementation
- ⏳ Adaptive questioning systematic
Then: Build Stages 2 & 3
User: "I want to redesign my office to make it more inspiring for work" Outcomist: Revealed real need was personal sanctuary with games, memorabilia, retreat space - not work optimization
User: "Should I take $6K/month client work or focus on my $2K MRR product?" Outcomist: Revealed $60K savings (12-14 months runway), product 4X growth, 97% retention. "Stable client" was $40/hour distraction. Recommended: Say no, go ALL IN on product.
User: "10 customers want Feature X, 2 customers want Feature Y. Which should I build?" Outcomist: Revealed product identity crisis - Feature X (workplace presence) serves employers for RTO, Feature Y (exclude list) serves employee safety. Can't be both. Recommended: Build Feature Y, clarify who you're building for.
Identify deeper pattern, reframe the question
Example:
You're framing this as "stable money vs bet on product" - but you have 12-14 months runway and your product grew 4X in 6 months with 97% retention. The scarcity is in your head, not your bank account.
Ask 4-6 targeted questions, deliver specific recommendation
Questions adapt based on answers - not a fixed script
For complex/high-stakes decisions, climb the ladder:
- Tier 3: Alternatives, risk analysis, scenarios (20-30 min)
- Tier 4: Comprehensive with stakeholder mapping (45-60 min)
Key: Only go deeper if complexity warrants it
- Progressive Disclosure: Start simple (30 sec), go deeper only if needed
- Honest Evidence: Pattern recognition yes, invented statistics no
- Adaptive Questioning: Questions evolve based on answers
- Clear Stance: "I recommend X" not "here are 7 options"
- One-by-One: Ask questions individually, wait for answers
This project lives inside the amplifier repository and has direct access to amplifier tools.
- Open this repository in GitHub
- Click "Code" → "Codespaces" → "Create codespace"
- The devcontainer automatically initializes the amplifier submodule
- Explore Command - Full command specification
- Pattern Library - Decision patterns
- Question Framework - Question heuristics
- Test Logs - Detailed test documentation
Version: 4.0 Last Updated: 2025-11-04