Thesis framing

Gate 1: Thesis Framing

Purpose: Articulate what we're deciding and why before any analysis.

Announce: "Starting Thesis Gate - let's clearly define the decision."

Entry Criteria

• Human has indicated a decision needs to be made
• No prior gates completed for this decision

Process

Ask questions one at a time to establish:

1. Decision Statement

• What exactly are we deciding?
• Frame as a clear yes/no or choice between alternatives
• "Should we [action]?" or "Which of [options] should we choose?"

2. Context & Trigger

• Why is this decision being made now?
• What happens if we don't decide?
• Is there a deadline?

3. Stakeholders

• Who is affected by this decision?
• Who has input? Who decides?
• Are there conflicting interests?

4. Constraints

• What limits the decision space?
• Budget, time, regulations, resources?
• Which constraints are non-negotiable vs flexible?

5. Success Criteria

• How will we know if this decision was right?
• What does success look like in 1 year? 5 years?
• What metrics matter?

6. Time Horizon & Reversibility

• One-way door or two-way door? (Bezos framing)
• How long until we know if it worked?
• What's the cost of reversal if wrong?

Depth by Weight

Aspect	Light	Medium	Complete
Questions	3-4 key questions	5-6 questions	Full exploration
Stakeholders	List only	Identify conflicts	Map relationships
Constraints	Non-negotiables only	All constraints	Constraint flexibility
Success criteria	1-2 metrics	3-4 metrics	Comprehensive + timeline

Light: Focus on decision statement, key constraints, reversibility. Skip deep stakeholder analysis.

Medium: Cover all sections at standard depth.

Complete: Explore each section thoroughly. Map stakeholder relationships. Analyze constraint flexibility.

Weight Suggestion

After completing Thesis Gate, suggest decision weight.

Analyze what you learned:

• Financial/strategic impact mentioned
• Reversibility assessment
• User's apparent domain expertise
• Complexity of stakeholder landscape
• Time pressure mentioned

Present weight suggestion:

Based on what you've described:
- [Key factor 1 - e.g., "$150K investment"]
- [Key factor 2 - e.g., "moderate reversibility (6 months to evaluate)"]
- [Key factor 3 - e.g., "you have domain expertise in this area"]

I recommend **[Weight]** depth (~[time]).

- **Light** (<10 min): Quick pass - good for reversible decisions where you have expertise
- **Medium** (30-40 min): Standard rigor - balances thoroughness with efficiency
- **Complete** (60-90 min): Full analysis - for high-stakes, hard-to-reverse decisions

Does [recommended weight] feel right, or would you prefer different depth?

Wait for user confirmation before proceeding.

If user pre-specified weight with /decide --light|medium|complete, skip this step.

Output

Create initial decision artifact:

# Decision: [Statement]

**Weight:** [Light/Medium/Complete]
**Estimated time:** [X minutes]

## Thesis Gate Summary

**Decision Statement:** [Clear framing]

**Trigger:** [Why now]

**Stakeholders:**
- Affected: [list]
- Input: [list]
- Decider: [name/role]

**Constraints:**
- Non-negotiable: [list]
- Flexible: [list]

**Success Criteria:** [metrics/outcomes]

**Reversibility:** [one-way/two-way, cost of reversal]

**Time Horizon:** [when we'll know]

Save to: docs/decisions/YYYY-MM-DD-<decision-slug>/decision.md

Exit Criteria

• Decision statement is clear and agreed
• Stakeholders identified
• Constraints documented
• Success criteria defined
• Reversibility assessed
• Weight confirmed by user

Bias Watch

Watch for:

• Framing effects - Is the question biased toward a particular answer?
• False dichotomy - Are there options we're not considering?
• Urgency bias - Is the deadline real or artificial?

Next Gate

Proceed to: deliberate-decisions:landscape-mapping