Name: evaluate-moats
Author: Zynapses

Prompt file imported from Zynapses/Radiant (.windsurf/workflows/evaluate-moats.md). Copyright stays with the author.

Evaluate Feature for Competitive Moat

MANDATORY: This workflow MUST be executed for EVERY new significant feature. No exceptions.

What Is a Moat?

A competitive moat is a feature or capability that:

Provides a real competitive advantage - Not available elsewhere
Is hard to replicate - Requires significant time/investment to copy
Creates switching costs - Customers lose value if they leave
Compounds over time - Gets stronger with usage

When to Trigger

This workflow is triggered for ANY new feature that is:

User-facing (affects customer experience)
Significant (not a bug fix or minor tweak)
Documented in an admin guide

Step 1: Initial Classification

Before scoring, determine the feature's classification:

✅ Elevate to Standalone Moat (Genuinely Novel)

Features that are:

Unprecedented - No competitor has anything similar
Demo-worthy - Impressive in investor/customer presentations
Paradigm-shifting - Changes how users think about AI tools

Examples: Reality Engine, Liquid Interface, Empiricism Loop

⚡ Consolidate into Existing Moat

Features that are:

Extensions of existing moat capabilities
Sub-components that strengthen a parent moat
Complementary services in a larger system

Feature Type	Consolidate Into
Learning/training services	Behavioral Learning System
Pattern extraction	Semantic Pattern Memory
Cost optimization	Unit Economics Advantage
Consciousness metrics	Genesis Cato Safety
Version/upgrade handling	AGI Brain / Ghost Vectors
Orchestration methods	Anti-Playbook Dynamic Reasoning

❌ Implementation Detail (Not a Moat)

Features that are:

Operational details - Good engineering, not strategic
Table stakes - Expected functionality competitors have
Cost optimizations - Merge into Unit Economics, not standalone
Standard practices - Circuit breakers, rate limiting, caching

Examples: Translation Middleware, Flash Facts, Persistence Guard, Admin Reports

Step 2: Moat Scoring Criteria

Score each criterion 1-5:

Criterion	Question	Score
Uniqueness	Does any competitor have this?	1=Common, 5=Only us
Replication Difficulty	How hard to copy?	1=Easy, 5=Very Hard
Data Network Effect	Does it get better with more users/data?	1=No, 5=Strong
Switching Cost	How painful to leave?	1=Easy, 5=Very Hard
Time Advantage	How long to catch up?	1=Days, 5=Years
Integration Depth	How deeply embedded?	1=Shallow, 5=Deep

Total Score: Add all scores (max 30)

Step 3: Tier Assignment

Total Score	Tier	Action
24-30	Tier 1: Technical	✅ Add as standalone moat
20-25	Tier 2: Architectural	✅ Add as standalone moat
18-22	Tier 3: Feature	⚠️ Add or consolidate (use judgment)
15-20	Tier 4: Business Model	⚠️ Add if business-model related
12-17	N/A	❌ NOT A MOAT - Document reason
6-11	N/A	❌ COMMODITY - Do not add

Current Moat Tiers

Tier	Count	Time to Replicate	Theme
Tier 1 (Technical)	8	18-24+ months	Autonomous Intelligence + Verifiable Truth
Tier 2 (Architectural)	8	12-18 months	Enterprise-Ready + Contextual Gravity
Tier 3 (Feature)	6	6-12 months	Market Gaps + Dynamic Reasoning
Tier 4 (Business)	3	3-9 months	Unit Economics + White-Label Strategy

Step 4: Documentation

Moat Documentation Targets

Moat Type	Primary Document	Secondary Document
RADIANT Platform (Infrastructure, Architecture)	`docs/15-STRATEGY-COMPETITIVE.md` (RADIANT Moats section)	`docs/15-STRATEGY-COMPETITIVE.md`
Think Tank (Consumer Features, UX)	`docs/15-STRATEGY-COMPETITIVE.md` (Think Tank Moats section)	`docs/15-STRATEGY-COMPETITIVE.md`
Cross-System (Affects both)	`docs/15-STRATEGY-COMPETITIVE.md` (both moat sections)	`docs/15-STRATEGY-COMPETITIVE.md`

For Standalone Moats (Score 18+)

Add to the appropriate moat section in docs/15-STRATEGY-COMPETITIVE.md
Add to the Competitive Strategy / Gap section in docs/15-STRATEGY-COMPETITIVE.md
Update the summary table in the moat section

RADIANT Moats section format (in docs/15-STRATEGY-COMPETITIVE.md):

### Moat #X: [Moat Name]

[Description - what it does and why it matters]

| Feature | Implementation |
|---------|----------------|
| ... | ... |

**Why It's a Moat**: [Why competitors can't replicate]

**Implementation**:
- Service: `lambda/shared/services/xxx.service.ts`
- Admin UI: `apps/admin-dashboard/...`

Think Tank Moats section format (in docs/15-STRATEGY-COMPETITIVE.md):

### Moat #X: [Moat Name]

[Description focused on user-facing benefit]

| Feature | Description |
|---------|-------------|
| ... | ... |

**Why It's a Moat**: [Why competitors can't replicate]

**Implementation**:
- Service: `lambda/thinktank/xxx.ts`
- Admin UI: `apps/thinktank-admin/...`

For Consolidated Features

Update the parent moat's description in docs/15-STRATEGY-COMPETITIVE.md to include the new capability.

For Non-Moats

Document in the feature's admin guide section:

**Moat Evaluation**: Score X/30 - Not a standalone moat. [Reason: operational detail / table stakes / consolidated into X]

When Features Are Modified or Deleted

MANDATORY: When a feature that is part of a moat is modified or deleted:

Update the corresponding moat section in docs/15-STRATEGY-COMPETITIVE.md
Update the Gap/Competitive Strategy section in docs/15-STRATEGY-COMPETITIVE.md if needed
Update implementation file paths if they changed
If moat is weakened significantly, re-evaluate its tier placement

Current 26 Moats Reference

Tier 1: Technical Moats (9)

Truth Engine™ (ECD Verification)
Genesis Cato Safety (Post-RLHF)
AGI Brain / Ghost Vectors
Self-Healing Reflexion Loop
Glass Box Auditability
Reality Engine (4 Superpowers)
Twilight Dreaming Cycle
Behavioral Learning System
Stub Nodes (Zero-Copy Data Gravity) - Score 27/30

Tier 2: Architectural Moats (8)

True Multi-Tenancy from Birth
Compliance Sandwich Architecture
Model-Agnostic Orchestration
Supply Chain Security
Contextual Gravity
Liquid Interface (50+ Components)
Tri-Layer LoRA Stacking
Empiricism Loop

Tier 3: Feature Moats (6)

Concurrent Task Execution
Real-Time Collaboration (Yjs CRDT)
Semantic Pattern Memory
Structure from Chaos Synthesis
Anti-Playbook Dynamic Reasoning
Curiosity Engine

Tier 4: Business Model Moats (3)

Unit Economics Advantage
Five Infrastructure Tiers
White-Label Invisibility

Examples of Consolidation Decisions

Feature	Decision	Rationale
Episode Logger	→ Behavioral Learning System	Part of 8-service learning loop
Paste-Back Detection	→ Behavioral Learning System	Strongest negative signal for learning
Recipe Extractor	→ Semantic Pattern Memory	Same concept—extracts patterns
Tool Entropy	→ Semantic Pattern Memory	Auto-chaining is pattern memory
Teacher-Student Distillation	→ Unit Economics Advantage	10x cost reduction is economic
IIT Phi Calculation	→ Genesis Cato Safety	Consciousness metrics
Version-Gated Upgrades	→ AGI Brain / Ghost Vectors	Already mentioned there
70+ Orchestration Methods	→ Anti-Playbook Dynamic Reasoning	Already covered

Examples of Non-Moat Decisions

Feature	Score	Reason
Translation Middleware	14/30	Operational detail, cost optimization
Semantic Blackboard	15/30	Agent coordination detail
Process Hydration	13/30	Technical implementation
Zero-Cost Ego	16/30	Merge into Unit Economics
Flash Facts	14/30	Reliability engineering
Magic Carpet Navigation	15/30	UX feature, not moat
Persistence Guard	12/30	Standard reliability
Semantic Cache	15/30	Merge into Unit Economics
Circuit Breakers	8/30	Table stakes
Admin Reports	10/30	Expected functionality
Dark Mode	6/30	Every competitor has it

Policy Enforcement

This evaluation is MANDATORY. Any significant feature PR/implementation that does not include a moat evaluation will be flagged for review.

The moat evaluation should be included in:

The CHANGELOG entry for the feature
The admin guide documentation
The Strategic Vision Marketing document (if qualified as moat)

Top 5 Demo-Ready Moats

When preparing investor presentations, prioritize these:

Rank	Moat	Demo Hook
1	Reality Engine	"Watch me time-travel to debug this code, then test across 3 parallel realities."
2	Liquid Interface	"This chat just became an app. Now exporting as deployable Next.js project."
3	Truth Engine	"Every dosage verified against sources. Watch the red flags when I hallucinate."
4	Concurrent Execution	"Running 4 AI models simultaneously, comparing and merging best parts."
5	Twilight Dreaming	"This deployment got 12% better overnight. AI learned while you slept."

Install evaluate-moats and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# Evaluate Feature for Competitive Moat **MANDATORY**: This workflow MUST be executed for EVERY new significant feature. No exceptions. ## What Is a Moat? A competitive moat is a feature or capability that: 1. Provides a **real competitive advantage** - Not available elsewhere 2. Is **hard to replicate** - Requires significant time/investment to copy 3. Creates **switching costs** - Customers lose value if they leave 4. **Compounds over time** - Gets stronger with usage ## When to Trigger This workflow is triggered for ANY new feature that is: - User-facing (affects customer experience) - Significant (not a bug fix or minor tweak) - Documented in an admin guide --- ## Step 1: Initial Classification Before scoring, determine the feature's classification: ### ✅ Elevate to Standalone Moat (Genuinely Novel) Features that are: - **Unprecedented** - No competitor has anything similar - **Demo-worthy** - Impressive in investor/customer presentations - **Paradigm-shifting** - Changes how users think about AI tools Examples: Reality Engine, Liquid Interface, Empiricism Loop ### ⚡ Consolidate into Existing Moat Features that are: - **Extensions** of existing moat capabilities - **Sub-components** that strengthen a parent moat - **Complementary** services in a larger system | Feature Type | Consolidate Into | |--------------|------------------| | Learning/training services | Behavioral Learning System | | Pattern extraction | Semantic Pattern Memory | | Cost optimization | Unit Economics Advantage | | Consciousness metrics | Genesis Cato Safety | | Version/upgrade handling | AGI Brain / Ghost Vectors | | Orchestration methods | Anti-Playbook Dynamic Reasoning | ### ❌ Implementation Detail (Not a Moat) Features that are: - **Operational details** - Good engineering, not strategic - **Table stakes** - Expected functionality competitors have - **Cost optimizations** - Merge into Unit Economics, not standalone - **Standard practices** - Circuit breakers, rate limiting, caching Examples: Translation Middleware, Flash Facts, Persistence Guard, Admin Reports --- ## Step 2: Moat Scoring Criteria Score each criterion 1-5: | Criterion | Question | Score | |-----------|----------|-------| | **Uniqueness** | Does any competitor have this? | 1=Common, 5=Only us | | **Replication Difficulty** | How hard to copy? | 1=Easy, 5=Very Hard | | **Data Network Effect** | Does it get better with more users/data? | 1=No, 5=Strong | | **Switching Cost** | How painful to leave? | 1=Easy, 5=Very Hard | | **Time Advantage** | How long to catch up? | 1=Days, 5=Years | | **Integration Depth** | How deeply embedded? | 1=Shallow, 5=Deep | **Total Score**: Add all scores (max 30) --- ## Step 3: Tier Assignment | Total Score | Tier | Action | |-------------|------|--------| | **24-30** | **Tier 1: Technical** | ✅ Add as standalone moat | | **20-25** | **Tier 2: Architectural** | ✅ Add as standalone moat | | **18-22** | **Tier 3: Feature** | ⚠️ Add or consolidate (use judgment) | | **15-20** | **Tier 4: Business Model** | ⚠️ Add if business-model related | | **12-17** | N/A | ❌ NOT A MOAT - Document reason | | **6-11** | N/A | ❌ COMMODITY - Do not add | ### Current Moat Tiers | Tier | Count | Time to Replicate | Theme | |------|-------|-------------------|-------| | **Tier 1 (Technical)** | 8 | 18-24+ months | Autonomous Intelligence + Verifiable Truth | | **Tier 2 (Architectural)** | 8 | 12-18 months | Enterprise-Ready + Contextual Gravity | | **Tier 3 (Feature)** | 6 | 6-12 months | Market Gaps + Dynamic Reasoning | | **Tier 4 (Business)** | 3 | 3-9 months | Unit Economics + White-Label Strategy | --- ## Step 4: Documentation ### Moat Documentation Targets | Moat Type | Primary Document | Secondary Document | |-----------|------------------|-------------------| | **RADIANT Platform** (Infrastructure, Architecture) | `docs/15-STRATEGY-COMPETITIVE.md` (RADIANT Moats section) | `docs/15-STRATEGY-COMPETITIVE.md` | | **Think Tank** (Consumer Features, UX) | `docs/15-STRATEGY-COMPETITIVE.md` (Think Tank Moats section) | `docs/15-STRATEGY-COMPETITIVE.md` | | **Cross-System** (Affects both) | `docs/15-STRATEGY-COMPETITIVE.md` (both moat sections) | `docs/15-STRATEGY-COMPETITIVE.md` | ### For Standalone Moats (Score 18+) 1. Add to the appropriate moat section in `docs/15-STRATEGY-COMPETITIVE.md` 2. Add to the Competitive Strategy / Gap section in `docs/15-STRATEGY-COMPETITIVE.md` 3. Update the summary table in the moat section **RADIANT Moats section format** (in `docs/15-STRATEGY-COMPETITIVE.md`): ```markdown ### Moat #X: [Moat Name] [Description - what it does and why it matters] | Feature | Implementation | |---------|----------------| | ... | ... | **Why It's a Moat**: [Why competitors can't replicate] **Implementation**: - Service: `lambda/shared/services/xxx.service.ts` - Admin UI: `apps/admin-dashboard/...` ``` **Think Tank Moats section format** (in `docs/15-STRATEGY-COMPETITIVE.md`): ```markdown ### Moat #X: [Moat Name] [Description focused on user-facing benefit] | Feature | Description | |---------|-------------| | ... | ... | **Why It's a Moat**: [Why competitors can't replicate] **Implementation**: - Service: `lambda/thinktank/xxx.ts` - Admin UI: `apps/thinktank-admin/...` ``` ### For Consolidated Features Update the parent moat's description in `docs/15-STRATEGY-COMPETITIVE.md` to include the new capability. ### For Non-Moats Document in the feature's admin guide section: ```markdown **Moat Evaluation**: Score X/30 - Not a standalone moat. [Reason: operational detail / table stakes / consolidated into X] ``` ### When Features Are Modified or Deleted **MANDATORY**: When a feature that is part of a moat is modified or deleted: 1. Update the corresponding moat section in `docs/15-STRATEGY-COMPETITIVE.md` 2. Update the Gap/Competitive Strategy section in `docs/15-STRATEGY-COMPETITIVE.md` if needed 3. Update implementation file paths if they changed 4. If moat is weakened significantly, re-evaluate its tier placement --- ## Current 26 Moats Reference ### Tier 1: Technical Moats (9) 1. Truth Engine™ (ECD Verification) 2. Genesis Cato Safety (Post-RLHF) 3. AGI Brain / Ghost Vectors 4. Self-Healing Reflexion Loop 5. Glass Box Auditability 6. Reality Engine (4 Superpowers) 7. Twilight Dreaming Cycle 8. Behavioral Learning System 9. Stub Nodes (Zero-Copy Data Gravity) - Score 27/30 ### Tier 2: Architectural Moats (8) 10. True Multi-Tenancy from Birth 11. Compliance Sandwich Architecture 12. Model-Agnostic Orchestration 13. Supply Chain Security 14. Contextual Gravity 15. Liquid Interface (50+ Components) 16. Tri-Layer LoRA Stacking 17. Empiricism Loop ### Tier 3: Feature Moats (6) 18. Concurrent Task Execution 19. Real-Time Collaboration (Yjs CRDT) 20. Semantic Pattern Memory 21. Structure from Chaos Synthesis 22. Anti-Playbook Dynamic Reasoning 23. Curiosity Engine ### Tier 4: Business Model Moats (3) 24. Unit Economics Advantage 25. Five Infrastructure Tiers 26. White-Label Invisibility --- ## Examples of Consolidation Decisions | Feature | Decision | Rationale | |---------|----------|-----------| | Episode Logger | → Behavioral Learning System | Part of 8-service learning loop | | Paste-Back Detection | → Behavioral Learning System | Strongest negative signal for learning | | Recipe Extractor | → Semantic Pattern Memory | Same concept—extracts patterns | | Tool Entropy | → Semantic Pattern Memory | Auto-chaining is pattern memory | | Teacher-Student Distillation | → Unit Economics Advantage | 10x cost reduction is economic | | IIT Phi Calculation | → Genesis Cato Safety | Consciousness metrics | | Version-Gated Upgrades | → AGI Brain / Ghost Vectors | Already mentioned there | | 70+ Orchestration Methods | → Anti-Playbook Dynamic Reasoning | Already covered | ## Examples of Non-Moat Decisions | Feature | Score | Reason | |---------|-------|--------| | Translation Middleware | 14/30 | Operational detail, cost optimization | | Semantic Blackboard | 15/30 | Agent coordination detail | | Process Hydration | 13/30 | Technical implementation | | Zero-Cost Ego | 16/30 | Merge into Unit Economics | | Flash Facts | 14/30 | Reliability engineering | | Magic Carpet Navigation | 15/30 | UX feature, not moat | | Persistence Guard | 12/30 | Standard reliability | | Semantic Cache | 15/30 | Merge into Unit Economics | | Circuit Breakers | 8/30 | Table stakes | | Admin Reports | 10/30 | Expected functionality | | Dark Mode | 6/30 | Every competitor has it | --- ## Policy Enforcement This evaluation is **MANDATORY**. Any significant feature PR/implementation that does not include a moat evaluation will be flagged for review. The moat evaluation should be included in: 1. The CHANGELOG entry for the feature 2. The admin guide documentation 3. The Strategic Vision Marketing document (if qualified as moat) --- ## Top 5 Demo-Ready Moats When preparing investor presentations, prioritize these: | Rank | Moat | Demo Hook | |------|------|-----------| | **1** | Reality Engine | "Watch me time-travel to debug this code, then test across 3 parallel realities." | | **2** | Liquid Interface | "This chat just became an app. Now exporting as deployable Next.js project." | | **3** | Truth Engine | "Every dosage verified against sources. Watch the red flags when I hallucinate." | | **4** | Concurrent Execution | "Running 4 AI models simultaneously, comparing and merging best parts." | | **5** | Twilight Dreaming | "This deployment got 12% better overnight. AI learned while you slept." |

{ "prompt": "# Evaluate Feature for Competitive Moat\n\n**MANDATORY**: This workflow MUST be executed for EVERY new significant feature. No exceptions.\n\n## What Is a Moat?\n\nA competitive moat is a feature or capability that:\n1. Provides a **real competitive advantage** - Not available elsewhere\n2. Is **hard to replicate** - Requires significant time/investment to copy\n3. Creates **switching costs** - Customers lose value if they leave\n4. **Compounds over time** - Gets stronger with usage\n\n## When to Trigger\n\nThis workflow is triggered for ANY new feature that is:\n- User-facing (affects customer experience)\n- Significant (not a bug fix or minor tweak)\n- Documented in an admin guide\n\n---\n\n## Step 1: Initial Classification\n\nBefore scoring, determine the feature's classification:\n\n### ✅ Elevate to Standalone Moat (Genuinely Novel)\n\nFeatures that are:\n- **Unprecedented** - No competitor has anything similar\n- **Demo-worthy** - Impressive in investor/customer presentations\n- **Paradigm-shifting** - Changes how users think about AI tools\n\nExamples: Reality Engine, Liquid Interface, Empiricism Loop\n\n### ⚡ Consolidate into Existing Moat\n\nFeatures that are:\n- **Extensions** of existing moat capabilities\n- **Sub-components** that strengthen a parent moat\n- **Complementary** services in a larger system\n\n| Feature Type | Consolidate Into |\n|--------------|------------------|\n| Learning/training services | Behavioral Learning System |\n| Pattern extraction | Semantic Pattern Memory |\n| Cost optimization | Unit Economics Advantage |\n| Consciousness metrics | Genesis Cato Safety |\n| Version/upgrade handling | AGI Brain / Ghost Vectors |\n| Orchestration methods | Anti-Playbook Dynamic Reasoning |\n\n### ❌ Implementation Detail (Not a Moat)\n\nFeatures that are:\n- **Operational details** - Good engineering, not strategic\n- **Table stakes** - Expected functionality competitors have\n- **Cost optimizations** - Merge into Unit Economics, not standalone\n- **Standard practices** - Circuit breakers, rate limiting, caching\n\nExamples: Translation Middleware, Flash Facts, Persistence Guard, Admin Reports\n\n---\n\n## Step 2: Moat Scoring Criteria\n\nScore each criterion 1-5:\n\n| Criterion | Question | Score |\n|-----------|----------|-------|\n| **Uniqueness** | Does any competitor have this? | 1=Common, 5=Only us |\n| **Replication Difficulty** | How hard to copy? | 1=Easy, 5=Very Hard |\n| **Data Network Effect** | Does it get better with more users/data? | 1=No, 5=Strong |\n| **Switching Cost** | How painful to leave? | 1=Easy, 5=Very Hard |\n| **Time Advantage** | How long to catch up? | 1=Days, 5=Years |\n| **Integration Depth** | How deeply embedded? | 1=Shallow, 5=Deep |\n\n**Total Score**: Add all scores (max 30)\n\n---\n\n## Step 3: Tier Assignment\n\n| Total Score | Tier | Action |\n|-------------|------|--------|\n| **24-30** | **Tier 1: Technical** | ✅ Add as standalone moat |\n| **20-25** | **Tier 2: Architectural** | ✅ Add as standalone moat |\n| **18-22** | **Tier 3: Feature** | ⚠️ Add or consolidate (use judgment) |\n| **15-20** | **Tier 4: Business Model** | ⚠️ Add if business-model related |\n| **12-17** | N/A | ❌ NOT A MOAT - Document reason |\n| **6-11** | N/A | ❌ COMMODITY - Do not add |\n\n### Current Moat Tiers\n\n| Tier | Count | Time to Replicate | Theme |\n|------|-------|-------------------|-------|\n| **Tier 1 (Technical)** | 8 | 18-24+ months | Autonomous Intelligence + Verifiable Truth |\n| **Tier 2 (Architectural)** | 8 | 12-18 months | Enterprise-Ready + Contextual Gravity |\n| **Tier 3 (Feature)** | 6 | 6-12 months | Market Gaps + Dynamic Reasoning |\n| **Tier 4 (Business)** | 3 | 3-9 months | Unit Economics + White-Label Strategy |\n\n---\n\n## Step 4: Documentation\n\n### Moat Documentation Targets\n\n| Moat Type | Primary Document | Secondary Document |\n|-----------|------------------|-------------------|\n| **RADIANT Platform** (Infrastructure, Architecture) | `docs/15-STRATEGY-COMPETITIVE.md` (RADIANT Moats section) | `docs/15-STRATEGY-COMPETITIVE.md` |\n| **Think Tank** (Consumer Features, UX) | `docs/15-STRATEGY-COMPETITIVE.md` (Think Tank Moats section) | `docs/15-STRATEGY-COMPETITIVE.md` |\n| **Cross-System** (Affects both) | `docs/15-STRATEGY-COMPETITIVE.md` (both moat sections) | `docs/15-STRATEGY-COMPETITIVE.md` |\n\n### For Standalone Moats (Score 18+)\n\n1. Add to the appropriate moat section in `docs/15-STRATEGY-COMPETITIVE.md`\n2. Add to the Competitive Strategy / Gap section in `docs/15-STRATEGY-COMPETITIVE.md`\n3. Update the summary table in the moat section\n\n**RADIANT Moats section format** (in `docs/15-STRATEGY-COMPETITIVE.md`):\n```markdown\n### Moat #X: [Moat Name]\n\n[Description - what it does and why it matters]\n\n| Feature | Implementation |\n|---------|----------------|\n| ... | ... |\n\n**Why It's a Moat**: [Why competitors can't replicate]\n\n**Implementation**:\n- Service: `lambda/shared/services/xxx.service.ts`\n- Admin UI: `apps/admin-dashboard/...`\n```\n\n**Think Tank Moats section format** (in `docs/15-STRATEGY-COMPETITIVE.md`):\n```markdown\n### Moat #X: [Moat Name]\n\n[Description focused on user-facing benefit]\n\n| Feature | Description |\n|---------|-------------|\n| ... | ... |\n\n**Why It's a Moat**: [Why competitors can't replicate]\n\n**Implementation**:\n- Service: `lambda/thinktank/xxx.ts`\n- Admin UI: `apps/thinktank-admin/...`\n```\n\n### For Consolidated Features\n\nUpdate the parent moat's description in `docs/15-STRATEGY-COMPETITIVE.md` to include the new capability.\n\n### For Non-Moats\n\nDocument in the feature's admin guide section:\n```markdown\n**Moat Evaluation**: Score X/30 - Not a standalone moat. [Reason: operational detail / table stakes / consolidated into X]\n```\n\n### When Features Are Modified or Deleted\n\n**MANDATORY**: When a feature that is part of a moat is modified or deleted:\n\n1. Update the corresponding moat section in `docs/15-STRATEGY-COMPETITIVE.md`\n2. Update the Gap/Competitive Strategy section in `docs/15-STRATEGY-COMPETITIVE.md` if needed\n3. Update implementation file paths if they changed\n4. If moat is weakened significantly, re-evaluate its tier placement\n\n---\n\n## Current 26 Moats Reference\n\n### Tier 1: Technical Moats (9)\n1. Truth Engine™ (ECD Verification)\n2. Genesis Cato Safety (Post-RLHF)\n3. AGI Brain / Ghost Vectors\n4. Self-Healing Reflexion Loop\n5. Glass Box Auditability\n6. Reality Engine (4 Superpowers)\n7. Twilight Dreaming Cycle\n8. Behavioral Learning System\n9. Stub Nodes (Zero-Copy Data Gravity) - Score 27/30\n\n### Tier 2: Architectural Moats (8)\n10. True Multi-Tenancy from Birth\n11. Compliance Sandwich Architecture\n12. Model-Agnostic Orchestration\n13. Supply Chain Security\n14. Contextual Gravity\n15. Liquid Interface (50+ Components)\n16. Tri-Layer LoRA Stacking\n17. Empiricism Loop\n\n### Tier 3: Feature Moats (6)\n18. Concurrent Task Execution\n19. Real-Time Collaboration (Yjs CRDT)\n20. Semantic Pattern Memory\n21. Structure from Chaos Synthesis\n22. Anti-Playbook Dynamic Reasoning\n23. Curiosity Engine\n\n### Tier 4: Business Model Moats (3)\n24. Unit Economics Advantage\n25. Five Infrastructure Tiers\n26. White-Label Invisibility\n\n---\n\n## Examples of Consolidation Decisions\n\n| Feature | Decision | Rationale |\n|---------|----------|-----------|\n| Episode Logger | → Behavioral Learning System | Part of 8-service learning loop |\n| Paste-Back Detection | → Behavioral Learning System | Strongest negative signal for learning |\n| Recipe Extractor | → Semantic Pattern Memory | Same concept—extracts patterns |\n| Tool Entropy | → Semantic Pattern Memory | Auto-chaining is pattern memory |\n| Teacher-Student Distillation | → Unit Economics Advantage | 10x cost reduction is economic |\n| IIT Phi Calculation | → Genesis Cato Safety | Consciousness metrics |\n| Version-Gated Upgrades | → AGI Brain / Ghost Vectors | Already mentioned there |\n| 70+ Orchestration Methods | → Anti-Playbook Dynamic Reasoning | Already covered |\n\n## Examples of Non-Moat Decisions\n\n| Feature | Score | Reason |\n|---------|-------|--------|\n| Translation Middleware | 14/30 | Operational detail, cost optimization |\n| Semantic Blackboard | 15/30 | Agent coordination detail |\n| Process Hydration | 13/30 | Technical implementation |\n| Zero-Cost Ego | 16/30 | Merge into Unit Economics |\n| Flash Facts | 14/30 | Reliability engineering |\n| Magic Carpet Navigation | 15/30 | UX feature, not moat |\n| Persistence Guard | 12/30 | Standard reliability |\n| Semantic Cache | 15/30 | Merge into Unit Economics |\n| Circuit Breakers | 8/30 | Table stakes |\n| Admin Reports | 10/30 | Expected functionality |\n| Dark Mode | 6/30 | Every competitor has it |\n\n---\n\n## Policy Enforcement\n\nThis evaluation is **MANDATORY**. Any significant feature PR/implementation that does not include a moat evaluation will be flagged for review.\n\nThe moat evaluation should be included in:\n1. The CHANGELOG entry for the feature\n2. The admin guide documentation\n3. The Strategic Vision Marketing document (if qualified as moat)\n\n---\n\n## Top 5 Demo-Ready Moats\n\nWhen preparing investor presentations, prioritize these:\n\n| Rank | Moat | Demo Hook |\n|------|------|-----------|\n| **1** | Reality Engine | \"Watch me time-travel to debug this code, then test across 3 parallel realities.\" |\n| **2** | Liquid Interface | \"This chat just became an app. Now exporting as deployable Next.js project.\" |\n| **3** | Truth Engine | \"Every dosage verified against sources. Watch the red flags when I hallucinate.\" |\n| **4** | Concurrent Execution | \"Running 4 AI models simultaneously, comparing and merging best parts.\" |\n| **5** | Twilight Dreaming | \"This deployment got 12% better overnight. AI learned while you slept.\" |", "variables": [], "notes": "MANDATORY - Evaluate every new significant feature for competitive moat status", "metadata": { "source": { "provider": "github-windsurf-workflows", "repository": "https://github.com/Zynapses/Radiant", "path": ".windsurf/workflows/evaluate-moats.md", "ref": "dfcafdfbbf2191ee48921207c5c8a8ca77da957b", "url": "https://github.com/Zynapses/Radiant/blob/dfcafdfbbf2191ee48921207c5c8a8ca77da957b/.windsurf/workflows/evaluate-moats.md", "key": "Zynapses/Radiant/.windsurf/workflows/evaluate-moats.md" } } }

evaluate-moats

About

Evaluate Feature for Competitive Moat

What Is a Moat?

When to Trigger

Step 1: Initial Classification

✅ Elevate to Standalone Moat (Genuinely Novel)

⚡ Consolidate into Existing Moat

❌ Implementation Detail (Not a Moat)

Step 2: Moat Scoring Criteria

Step 3: Tier Assignment

Current Moat Tiers

Step 4: Documentation

Moat Documentation Targets

For Standalone Moats (Score 18+)

For Consolidated Features

For Non-Moats

When Features Are Modified or Deleted

Current 26 Moats Reference

Tier 1: Technical Moats (9)

Tier 2: Architectural Moats (8)

Tier 3: Feature Moats (6)

Tier 4: Business Model Moats (3)

Examples of Consolidation Decisions

Examples of Non-Moat Decisions

Policy Enforcement

Top 5 Demo-Ready Moats

Install evaluate-moats and it becomes one more candidate the router can pick - when it fits.

Use it

Manifest

Reviews