ultra-builder-pro-cli 0.5.0

package/skills/ultra-research/steps/step-03-competitive-landscape.md

@@ -0,0 +1,197 @@
+# Step 03: Competitive Landscape
+
+## MANDATORY EXECUTION RULES
+
+- WEB SEARCH REQUIRED — identify real competitors, not imagined ones
+- INCLUDE "DO NOTHING" — the biggest competitor is often the status quo
+- ANALYZE 2-5 DIRECT competitors + 1-2 indirect alternatives
+- BE HONEST about where competitors are stronger
+- WRITE output to spec file BEFORE presenting to user
+- ALL output in English (spec files); conversation in Chinese
+
+## PREREQUISITES
+
+- Step 02 (Market Assessment) completed with [C]
+- `.ultra/specs/discovery.md` §0-2 exist
+
+## CONTEXT BOUNDARIES
+
+- Focus: Who else is solving this problem? How do we differentiate?
+- Use the prioritized opportunities from step-01 to frame competitive analysis
+- Analyze CURRENT competitors, not hypothetical future ones
+- Include indirect competitors and the "do nothing" alternative
+
+## SEARCH STRATEGY (MANDATORY)
+
+Execute these web searches in parallel:
+
+```
+Search: "{{product_domain}} competitors comparison review"
+Search: "{{product_domain}} alternatives tools software"
+Search: "{{product_domain}} market leaders market share"
+Search: "{{product_domain}} startup funding competitors"
+Search: "{{product_domain}} vs [known competitor if any]"
+```
+
+For each identified competitor, execute:
+```
+Search: "[competitor name] pricing features review"
+Search: "[competitor name] weaknesses complaints"
+```
+
+## EXECUTION SEQUENCE
+
+### 1. Identify Competitive Players
+
+Categorize all competitors found:
+
+**Direct Competitors**: Products solving the same problem for the same segment
+- Search for 2-5 direct competitors
+- Include their funding, team size, and traction if available
+
+**Indirect Competitors / Alternatives**: What customers use today instead
+- Spreadsheets, manual processes, general-purpose tools
+- Adjacent products that partially solve the problem
+
+**"Do Nothing" Alternative**: What happens if the customer ignores the problem?
+- What does inaction cost? (Reference step-00 status quo analysis)
+- Why do some users choose to live with the pain?
+
+### 2. Deep-Dive Each Competitor
+
+For each direct competitor (2-5), research:
+
+| Dimension | What to find |
+|-----------|-------------|
+| Core value prop | What do they promise? |
+| Target segment | Who do they serve? |
+| Pricing model | How do they charge? |
+| Key strengths | What are they genuinely good at? |
+| Key weaknesses | Where do users complain? (check reviews, forums) |
+| Traction signals | Users, revenue, funding, growth |
+| Defensibility | What makes them hard to beat? |
+
+### 3. Porter's Five Forces (Brief)
+
+Assess competitive intensity:
+
+| Force | Rating (1-5) | Assessment |
+|-------|-------------|-----------|
+| Supplier power | | [Who controls key inputs?] |
+| Buyer power | | [Can customers easily switch?] |
+| New entrant threat | | [How easy to enter this market?] |
+| Substitute threat | | [What alternatives exist?] |
+| Rivalry intensity | | [How fierce is current competition?] |
+
+### 4. Identify Our Competitive Edge
+
+Based on the analysis, answer:
+- Where can we win? (specific dimensions where we have an advantage)
+- Where can we NOT win? (be honest — avoid head-to-head on competitor strengths)
+- What is the gap that no one is filling?
+
+### 5. Write Output
+
+**WRITE IMMEDIATELY** to `.ultra/specs/discovery.md` §3:
+
+```markdown
+## §3 Competitive Landscape
+
+### Competitive Overview
+
+| Dimension | Us | [Competitor A] | [Competitor B] | [Competitor C] | Do Nothing |
+|-----------|-----|---------------|---------------|---------------|------------|
+| Core value prop | [ours] | [theirs] | [theirs] | [theirs] | [status quo] |
+| Target segment | [ours] | [theirs] | [theirs] | [theirs] | [everyone] |
+| Pricing | [ours] | [theirs] | [theirs] | [theirs] | Free |
+| Key strength | [ours] | [theirs] | [theirs] | [theirs] | Familiar |
+| Key weakness | [ours] | [theirs] | [theirs] | [theirs] | Inefficient |
+| Traction | [ours] | [theirs] | [theirs] | [theirs] | N/A |
+
+### Direct Competitors
+
+#### [Competitor A]
+- **What they do**: [Description]
+- **Strengths**: [What they're genuinely good at]
+- **Weaknesses**: [Where users complain] — Source: [URL]
+- **Pricing**: [Model and price points] — Source: [URL]
+- **Traction**: [Users/revenue/funding] — Source: [URL]
+- **Threat level**: [High / Medium / Low] — [Why]
+
+#### [Competitor B]
+[Same structure]
+
+#### [Competitor C]
+[Same structure]
+
+### Indirect Competitors & Alternatives
+- **[Tool/Process A]**: [How it partially solves the problem]
+- **[Tool/Process B]**: [How it partially solves the problem]
+- **Do Nothing**: [Cost of inaction from step-00]
+
+### Porter's Five Forces Summary
+
+| Force | Rating | Assessment |
+|-------|--------|-----------|
+| Supplier power | [1-5] | [Brief assessment] |
+| Buyer power | [1-5] | [Brief assessment] |
+| New entrant threat | [1-5] | [Brief assessment] |
+| Substitute threat | [1-5] | [Brief assessment] |
+| Rivalry intensity | [1-5] | [Brief assessment] |
+| **Overall competitive intensity** | **[1-5]** | **[Summary]** |
+
+### Our Competitive Edge
+- **Where we can win**: [Specific dimensions with reasoning]
+- **Where we cannot win**: [Honest assessment of competitor strengths]
+- **Unfilled gap**: [What no one is doing that we can do]
+- **Timing advantage**: [Why now is the right time]
+
+### Competitive Confidence
+- **Overall confidence**: [X]%
+- **Best-researched competitor**: [Name] — [depth of analysis]
+- **Blind spot**: [What we might be missing]
+```
+
+### 6. Present to User and Gate
+
+Show the Competitive Landscape summary. Highlight:
+- The most dangerous competitor and why
+- The clearest gap in the market
+- Any surprising findings (e.g., a competitor we didn't expect)
+
+```
+[C] Continue — Competitive landscape mapped, proceed to Product Strategy
+[R] Revise — Adjust competitor analysis or add missing competitors
+[S] Search more — Deep-dive a specific competitor
+```
+
+**HALT — wait for user response before proceeding.**
+
+### 7. Handle Response
+
+- **[C]**: Load next step: `./step-04-product-strategy.md`
+- **[R]**: Revise analysis, update discovery.md §3, re-present
+- **[S]**: Execute targeted searches on specific competitor, update, re-present
+
+## SUCCESS METRICS
+
+- 2-5 direct competitors identified with evidence
+- "Do nothing" alternative analyzed
+- Comparison matrix completed for all competitors
+- Porter's Five Forces assessed
+- Honest assessment of where we can/cannot win
+- All claims have source citations
+- Output written to discovery.md §3
+
+## FAILURE MODES
+
+- "No competitors exist" (there are ALWAYS competitors, even if indirect)
+- Only analyzing competitor strengths, ignoring weaknesses
+- Only analyzing competitor weaknesses, ignoring strengths (cheerleading)
+- Missing the "do nothing" alternative
+- No source citations for competitor data
+- Not writing output before presenting to user
+
+## NEXT STEP
+
+After user selects [C], read and follow: `./step-04-product-strategy.md`

package/skills/ultra-research/steps/step-04-product-strategy.md

@@ -0,0 +1,196 @@
+# Step 04: Product Strategy
+
+## MANDATORY EXECUTION RULES
+
+- STRATEGY IS ABOUT TRADE-OFFS — what we choose NOT to do matters more
+- REFERENCE previous steps — strategy must be grounded in validated problem + market + competition
+- WEB SEARCH for comparable strategies in the domain
+- WRITE output to spec file BEFORE presenting to user
+- ALL output in English (spec files); conversation in Chinese
+
+## PREREQUISITES
+
+- Steps 00-03 completed with [C]
+- `.ultra/specs/discovery.md` §0-3 exist
+
+## CONTEXT BOUNDARIES
+
+- Focus: Define strategic direction — vision, segments, value prop, trade-offs, defensibility
+- This is a CONDENSED strategy canvas, not a full business plan
+- Must be grounded in evidence from previous steps
+- Strategy should be opinionated — "we serve everyone" is not a strategy
+
+## SEARCH STRATEGY (MANDATORY)
+
+Execute these web searches:
+
+```
+Search: "{{product_domain}} product strategy examples successful"
+Search: "{{product_domain}} go-to-market strategy startup"
+Search: "{{product_domain}} defensibility moat competitive advantage"
+Search: "{{product_domain}} pricing strategy models"
+```
+
+## EXECUTION SEQUENCE
+
+### 1. Vision Statement
+
+Collaborate with user to craft a 2-3 sentence vision that:
+- Inspires people (emotional, memorable)
+- Is specific enough to guide decisions
+- Is ambitious enough to attract talent and investment
+
+**Test**: Would someone quit their job to work on this? If not, it's too boring.
+
+Ask user via AskUserQuestion: "In 2-3 sentences, how can we inspire people? What are we aspiring to achieve?"
+
+### 2. Target Segments
+
+Define WHO we serve and WHO we explicitly do NOT serve.
+
+Using personas from step-00 (Q3: Desperate Specificity):
+
+**Serve** (defined by problems/JTBD, not demographics):
+- Segment 1: [Description] — [Why they need us most]
+- Segment 2: [Description] — [Why they need us]
+
+**Do NOT serve** (equally important):
+- Anti-segment 1: [Description] — [Why not]
+- Anti-segment 2: [Description] — [Why not]
+
+### 3. Value Proposition
+
+For each target segment, use the JTBD format:
+
+```
+When [situation], they want [motivation], so they can [outcome]
+```
+
+This must connect directly to the top opportunities from step-01.
+
+### 4. Strategic Trade-offs
+
+**This is the most important section.** A strategy without trade-offs is a wish list.
+
+Identify 3-5 strategic trade-offs:
+
+| We Choose | Over | Because |
+|-----------|------|---------|
+| [Focus A] | [Alternative A] | [Reasoning grounded in evidence] |
+| [Focus B] | [Alternative B] | [Reasoning grounded in evidence] |
+| [Focus C] | [Alternative C] | [Reasoning grounded in evidence] |
+
+Each trade-off should reference competitive analysis from step-03 — we avoid competing head-to-head on competitor strengths.
+
+### 5. Defensibility Analysis
+
+What makes this hard to copy? Evaluate each:
+
+| Moat Type | Applicability | Strength | Timeline to Build |
+|-----------|-------------|----------|-------------------|
+| Network effects | [Yes/No] | [Assessment] | [Time] |
+| Data advantage | [Yes/No] | [Assessment] | [Time] |
+| Switching costs | [Yes/No] | [Assessment] | [Time] |
+| Brand / trust | [Yes/No] | [Assessment] | [Time] |
+| Technical IP | [Yes/No] | [Assessment] | [Time] |
+| Speed / execution | [Yes/No] | [Assessment] | [Time] |
+
+Be honest — most startups have weak defensibility early. That's okay if the execution advantage is strong.
+
+### 6. Write Output
+
+**WRITE IMMEDIATELY** to `.ultra/specs/discovery.md` §4:
+
+```markdown
+## §4 Product Strategy
+
+### Vision
+[2-3 sentences — inspiring, specific, ambitious]
+
+### Target Segments
+
+#### Primary: [Segment Name]
+- **Description**: [Who they are, defined by problem/JTBD]
+- **Why them**: [Why they need us most urgently]
+- **Value Proposition**: When [situation], they want [motivation], so they can [outcome]
+
+#### Secondary: [Segment Name]
+- **Description**: [Who they are]
+- **Why them**: [Why they need us]
+- **Value Proposition**: When [situation], they want [motivation], so they can [outcome]
+
+#### Explicitly NOT Serving
+- **[Anti-segment 1]**: [Why not — e.g., "Too enterprise for our current capabilities"]
+- **[Anti-segment 2]**: [Why not — e.g., "Their problem is different enough to require a different product"]
+
+### Strategic Trade-offs
+
+| # | We Choose | Over | Because |
+|---|-----------|------|---------|
+| 1 | [Focus] | [Alternative] | [Evidence-based reasoning] |
+| 2 | [Focus] | [Alternative] | [Evidence-based reasoning] |
+| 3 | [Focus] | [Alternative] | [Evidence-based reasoning] |
+| 4 | [Focus] | [Alternative] | [Evidence-based reasoning] |
+
+### Defensibility
+
+| Moat Type | Applicable | Strength (1-5) | Timeline |
+|-----------|-----------|----------------|----------|
+| Network effects | [Yes/No] | [X] | [Time] |
+| Data advantage | [Yes/No] | [X] | [Time] |
+| Switching costs | [Yes/No] | [X] | [Time] |
+| Brand / trust | [Yes/No] | [X] | [Time] |
+| Technical IP | [Yes/No] | [X] | [Time] |
+| Speed / execution | [Yes/No] | [X] | [Time] |
+
+**Primary moat**: [Which moat type is strongest and why]
+**Moat timeline**: [When defensibility becomes meaningful]
+
+### Strategy Confidence
+- **Overall confidence**: [X]%
+- **Strongest element**: [What part of strategy is most grounded]
+- **Riskiest bet**: [What strategic choice is most uncertain]
+```
+
+### 7. Present to User and Gate
+
+Show the Product Strategy summary. Highlight:
+- Whether the trade-offs feel right
+- The defensibility assessment
+- Any concerns about the strategic direction
+
+```
+[C] Continue — Strategy defined, proceed to Assumptions Validation
+[R] Revise — Adjust strategy elements
+[D] Discuss — Explore specific trade-offs in more depth
+```
+
+**HALT — wait for user response before proceeding.**
+
+### 8. Handle Response
+
+- **[C]**: Load next step: `./step-05-assumptions-validation.md`
+- **[R]**: Revise strategy, update discovery.md §4, re-present
+- **[D]**: Deep-dive specific trade-offs, then re-present gate
+
+## SUCCESS METRICS
+
+- Vision is inspiring and specific (not generic)
+- Target segments defined by problems, not demographics
+- Anti-segments explicitly stated
+- At least 3 strategic trade-offs with evidence-based reasoning
+- Defensibility honestly assessed
+- Output written to discovery.md §4
+
+## FAILURE MODES
+
+- Generic vision ("We make the world better with AI")
+- "We serve everyone" (no segment focus)
+- No trade-offs stated (strategy without trade-offs is not strategy)
+- Defensibility assessment is all 5/5 (unrealistic)
+- Strategy not grounded in evidence from steps 00-03
+- Not writing output before presenting to user
+
+## NEXT STEP
+
+After user selects [C], read and follow: `./step-05-assumptions-validation.md`

package/skills/ultra-research/steps/step-05-assumptions-validation.md

@@ -0,0 +1,234 @@
+# Step 05: Assumptions & Validation Plan
+
+## MANDATORY EXECUTION RULES
+
+- EXTRACT assumptions from ALL previous steps (00-04) — not just the obvious ones
+- PRIORITIZE by Impact × Uncertainty — focus on "leap of faith" assumptions
+- DESIGN cheap, fast experiments — not "build it and see"
+- WRITE output to spec file BEFORE presenting to user
+- ALL output in English (spec files); conversation in Chinese
+
+## PREREQUISITES
+
+- Steps 00-04 completed with [C]
+- `.ultra/specs/discovery.md` §0-4 exist
+
+## CONTEXT BOUNDARIES
+
+- Focus: What are we assuming? What's riskiest? How do we validate cheaply?
+- This is the LAST step of Round 0 — it synthesizes everything discovered so far
+- After this step, user decides: proceed to Round 1 or investigate further
+- Based on Alberto Savoia (Pretotyping) and Marty Cagan methodologies
+
+## SEARCH STRATEGY (MANDATORY)
+
+Execute these web searches:
+
+```
+Search: "{{product_domain}} validation methods pretotype MVP"
+Search: "{{product_domain}} common startup mistakes failed assumptions"
+Search: "landing page test fake door validation {{product_domain}}"
+Search: "{{product_domain}} customer interview validation techniques"
+```
+
+## EXECUTION SEQUENCE
+
+### 1. Extract Assumptions from Previous Steps
+
+Review §0-4 and extract assumptions across 5 categories:
+
+| Category | Source | What to look for |
+|----------|--------|-----------------|
+| **Value** | §0, §1 | Will users actually want this? |
+| **Usability** | §1 | Can users figure it out without help? |
+| **Feasibility** | §1 (solutions) | Can we build it with current technology? |
+| **Viability** | §2 | Does the business case work at these numbers? |
+| **Go-to-Market** | §3, §4 | Can we reach and convert our target users? |
+
+### 2. Prioritize Assumptions
+
+Map each assumption on Impact × Uncertainty:
+
+```
+                    High Impact
+                        │
+    VALIDATE FIRST ─────┼───── MONITOR
+    (Leap of Faith)     │     (Important but clear)
+                        │
+   Low Uncertainty ─────┼───── High Uncertainty
+                        │
+    IGNORE ─────────────┼───── INVESTIGATE
+    (Low stakes)        │     (Uncertain but low impact)
+                        │
+                    Low Impact
+```
+
+Focus on "Leap of Faith" quadrant: **High Impact + High Uncertainty**
+
+### 3. Design Validation Experiments
+
+For top 3-5 leap-of-faith assumptions, design cheap, fast experiments:
+
+**Experiment Design Framework:**
+
+| Element | Description |
+|---------|-------------|
+| Assumption | What exactly are we assuming? |
+| Category | Value / Usability / Feasibility / Viability / GTM |
+| Method | How will we test this? |
+| Success criteria | What result validates the assumption? |
+| Failure criteria | What result invalidates it? |
+| Effort | Hours/days to execute |
+| Timeline | When can we have results? |
+
+**Preferred Methods** (cheapest first):
+1. **Data analysis**: Existing data that proves/disproves the assumption
+2. **Customer interviews**: 5-10 targeted conversations
+3. **Fake door test**: Landing page with sign-up / "notify me" button
+4. **Concierge MVP**: Manually deliver the value to 3-5 users
+5. **Pretotype**: Mechanical Turk version of the product
+6. **Prototype**: Clickable mockup tested with real users
+7. **MVP**: Minimum viable version with real functionality
+
+### 4. Define Decision Framework
+
+For each experiment:
+
+```
+If experiment SUCCEEDS → [Next action]
+If experiment FAILS → [Pivot / Investigate / Kill]
+If results are AMBIGUOUS → [How to get clarity]
+```
+
+### 5. Write Output
+
+**WRITE IMMEDIATELY** to `.ultra/specs/discovery.md` §5:
+
+```markdown
+## §5 Key Assumptions & Validation Plan
+
+### Assumption Inventory
+
+| # | Assumption | Category | Impact | Uncertainty | Priority |
+|---|-----------|----------|--------|-------------|----------|
+| A1 | [Statement] | Value | High | High | **Validate First** |
+| A2 | [Statement] | Viability | High | High | **Validate First** |
+| A3 | [Statement] | GTM | High | Medium | **Validate First** |
+| A4 | [Statement] | Feasibility | Medium | High | Investigate |
+| A5 | [Statement] | Usability | Medium | Medium | Monitor |
+| A6 | [Statement] | Value | Low | Low | Ignore |
+
+### Leap-of-Faith Assumptions (Top 3-5)
+
+#### A1: [Assumption Statement]
+- **Category**: [Value / Viability / GTM / Feasibility / Usability]
+- **Why it matters**: [What happens if this is wrong]
+- **Current evidence**: [What we know from steps 00-04]
+- **Evidence gap**: [What we don't know]
+
+**Validation Experiment:**
+- **Method**: [Specific method]
+- **Success criteria**: [Measurable outcome that validates]
+- **Failure criteria**: [Measurable outcome that invalidates]
+- **Effort**: [Hours/days]
+- **Timeline**: [When results expected]
+
+**Decision Framework:**
+- ✅ If validated → [Proceed to Round 1]
+- ❌ If invalidated → [Pivot to X / Investigate Y / Kill]
+- ⚠️ If ambiguous → [Additional experiment Z]
+
+#### A2: [Assumption Statement]
+[Same structure]
+
+#### A3: [Assumption Statement]
+[Same structure]
+
+### Validation Roadmap
+
+| Week | Experiment | Assumption | Expected Result |
+|------|-----------|-----------|----------------|
+| 1 | [Experiment 1] | A1 | [What we'll learn] |
+| 1-2 | [Experiment 2] | A2 | [What we'll learn] |
+| 2-3 | [Experiment 3] | A3 | [What we'll learn] |
+
+### Round 0 Summary
+
+**Product Discovery Confidence**:
+- **Problem validation**: [X]% — [One-line summary]
+- **Opportunity space**: [X]% — [One-line summary]
+- **Market size**: [X]% — [One-line summary]
+- **Competitive position**: [X]% — [One-line summary]
+- **Strategy clarity**: [X]% — [One-line summary]
+- **Overall R0 confidence**: [X]%
+
+**Recommendation**: [Proceed to Round 1 / Validate assumptions first / Pivot]
+```
+
+### 6. Write Round 0 Research Report
+
+**WRITE** to `.ultra/docs/research/product-discovery-{date}.md`:
+
+```markdown
+# Round 0: Product Discovery & Strategy
+
+> **Confidence**: [X]%
+> **Steps completed**: 00-05
+> **Completed**: [date]
+
+## Key Findings
+[3-5 bullet points of most important discoveries]
+
+## Decisions Made
+[Strategic choices and their rationale]
+
+## Open Questions
+[What remains uncertain]
+
+## Assumptions to Validate
+[Top 3 from §5]
+```
+
+### 7. Present to User and Gate
+
+Show the complete Round 0 summary. Highlight:
+- The top 3 leap-of-faith assumptions
+- Overall confidence level
+- Recommendation (proceed / validate / pivot)
+
+```
+[C] Continue — Proceed to Round 1 (User & Scenario Discovery)
+[V] Validate — Pause to run validation experiments before continuing
+[P] Pivot — Rethink the product direction based on findings
+```
+
+**HALT — wait for user response before proceeding.**
+
+### 8. Handle Response
+
+- **[C]**: Load next step: `./step-10-user-personas.md`
+- **[V]**: Help design validation experiments in detail, then re-present gate when done
+- **[P]**: Restart from step-00 with revised direction
+
+## SUCCESS METRICS
+
+- Assumptions extracted from ALL previous steps (not just surface-level)
+- Each assumption categorized and prioritized
+- Top 3-5 have detailed validation experiments
+- Each experiment has clear success/failure criteria
+- Decision framework defined for each outcome
+- Round 0 research report written
+- Output written to discovery.md §5
+
+## FAILURE MODES
+
+- Only extracting obvious assumptions (missing viability, GTM)
+- Designing expensive experiments ("build an MVP") when cheaper options exist
+- No clear success/failure criteria for experiments
+- Missing the decision framework (what do we DO with the results?)
+- Overly optimistic confidence assessment
+- Not writing both spec file AND research report
+
+## NEXT STEP
+
+After user selects [C], read and follow: `./step-10-user-personas.md`