foundation-builder
Orchestrate 10 agents to build and maintain validated Lean Canvas foundation. Handles business model mode selection, market discovery (mode-aware), validation, business model, and GTM. Pre-launch population and operational maintenance.
- Category: discovered
- Tags: github, community
- Version: 1.0.0
- Author: walid
- Downloads: 0
- GitHub stars: 3
- Updated: 2025-11-21
Download the latest version (v1.0.0) from this page.
SKILL.md
---
name: foundation-builder
description: Orchestrate 10 agents to build and maintain validated Lean Canvas foundation. Handles business model mode selection, market discovery (mode-aware), validation, business model, and GTM. Pre-launch population and operational maintenance.
allowed-tools: "Read,Write,Bash"
---
# Foundation Builder: Canvas Population & Validation
## Overview
Business foundation orchestration system that coordinates 10 specialized agents to build and maintain a validated Lean Canvas - the single source of truth for all strategic decisions.
**Primary Mission:** Populate and validate all 16 Canvas sections with evidence-based information (including business model mode selection)
**Lifecycle:**
- Pre-launch: Build foundation (Discovery → Launch)
- Operational: Maintain foundation (Canvas updates from validated learning)
**Target Profile**: AI-first lean startups operating with minimal human oversight
**Final Output:** Living Canvas that continuously reflects current business reality
---
## System Architecture
### Agent Organization
The system comprises 10 specialized agents organized into two tiers:
**Core Agents (Always Active - 7 agents)**
1. **market-intelligence**: Market analysis, competitive landscape, customer segmentation
2. **problem-solution-fit**: Problem validation, solution design, MVP definition
3. **value-proposition**: Unique value articulation, positioning, messaging
4. **business-model**: Revenue design, cost structure, unit economics
5. **validation**: Assumption testing, experiment design, hypothesis validation
6. **go-to-market**: Channel strategy, launch planning, growth tactics
7. **execution**: Task orchestration using Claude/GPT/Copilot, daily operations
**Specialist Agents (On-Demand - 3 agents)**
8. **funding**: Fundraising strategy, pitch development, investor targeting
9. **regulatory**: Compliance requirements, licensing, regulated market entry
10. **retention-optimizer**: Cohort analysis, churn prediction, optimization experiments
### Canvas Population Flow
Each agent populates specific Canvas sections:
| Agent | Canvas Sections |
|-------|----------------|
| **market-intelligence** | 01-context (KBOS), 02-constraints, 03-opportunity, 04-segments, 06-competitive |
| **problem-solution-fit** | 05-problem, 09-solution, 10-assumptions |
| **value-proposition** | 07-uvp, 08-advantage |
| **business-model** | 11-pricing, 12-costs, 13-metrics |
| **validation** | 10-assumptions (validation status) |
| **go-to-market** | 14-growth, 15-gtm |
| **funding** | `strategy/financials/fundraising.md` |
### Information Flow
The system follows a structured progression through pre-launch phases:
```
Mode Discovery → Definition → Validation → Execution → Launch
Select ↓ ↓ ↓ ↓ ↓
↓ Market Value Prop Validation Execution GTM
↓ Intel → + Business → Agent → Agent → Agent
↓ Agent Model
↓ ↓ ↓ ↓ ↓ ↓
Canvas Canvas Canvas Canvas artifacts/ Canvas
00 01-06 07-12 10 (val) Outputs 14-15
Mode determines: Which market research skill (venture vs bootstrap)
```
---
## Orchestration Logic
### Phase Detection
Determine the current pre-launch phase by analyzing the request context:
**Mode Selection Phase (0)**
- Triggers: Initial setup, business model strategy questions, "which mode should I use"
- Action: Create `00-business-model-mode.md` with VENTURE/BOOTSTRAP/HYBRID declaration
- Output Focus: Business model mode declaration and rationale
- Canvas Sections: 00-business-model-mode
- **CRITICAL: This must be completed before Discovery Phase**
**Discovery Phase (0a) - MODE-AWARE**
- Triggers: Market research, TAM/SAM/SOM questions, competitive analysis, customer segmentation
- **BEFORE ROUTING:** Check `strategy/canvas/00-business-model-mode.md` for mode
- Primary Agents:
- **If VENTURE:** market-intelligence (TAM focus) + market-research-venture skill
- **If BOOTSTRAP:** market-intelligence (spend mapping) + market-research-bootstrap skill
- **Both:** problem-solution-fit
- Output Focus: Mode-appropriate market understanding, problem identification
- Canvas Sections: 00-business-model-mode (if not exists), 01-context, 02-constraints, 03-opportunity, 04-segments, 05-problem, 06-competitive
**Definition Phase (0b)**
- Triggers: Product strategy, value proposition, business model design, pricing strategy
- Primary Agents: value-proposition, business-model, problem-solution-fit
- Output Focus: Clear positioning, revenue model, MVP scope
- Canvas Sections: 07-uvp, 08-advantage, 09-solution, 11-pricing, 12-costs, 13-metrics
**Validation Phase (1a)**
- Triggers: Testing assumptions, experiment design, MVP validation, pivot decisions
- Primary Agents: validation, execution
- Output Focus: Validated hypotheses, experiment results, pivot triggers
- Canvas Sections: 10-assumptions (status updates)
**Execution Phase (1b)**
- Triggers: Feature development, sprint planning, MVP build, implementation work
- Primary Agents: execution, validation
- Output Focus: Completed MVP, shipped code, progress metrics
- Outputs: Code, `artifacts/` deliverables
**Launch Phase (1c)**
- Triggers: Launch planning, channel strategy, early user acquisition
- Primary Agents: go-to-market, execution
- Output Focus: Launch playbook, initial traction, first customers
- Canvas Sections: 14-growth, 15-gtm
- **Completion:** Once you have first customers + validated Canvas → pre-launch work complete
**Specialist Scenarios**
- Fundraising: funding agent (Seed preparation)
- Compliance: regulatory agent (regulated markets, licensing)
- Retention Analysis: retention-optimizer agent (post-first customers)
### Routing Strategy
When a request arrives, follow this decision process:
1. **Check Business Model Mode**:
- Read `strategy/canvas/00-business-model-mode.md`
- If doesn't exist → Create it first (Phase 0: Mode Selection)
- Mode determines market research approach and decision criteria
2. **Identify Current Phase**: Determine where in pre-launch lifecycle (0 → 0a → 1c)
3. **Detect Specialist Needs**: Check if funding, regulatory, or retention scenarios apply
4. **Select Primary Agent(s)**: Choose 1-2 agents that best match the request
- **For market research:** Use mode-appropriate skill
- VENTURE: market-research-venture (TAM, growth, scale)
- BOOTSTRAP: market-research-bootstrap (spend, arbitrage, profit)
5. **Identify Canvas Target**: Which Canvas sections will be populated/updated?
6. **Coordinate Multi-Agent Workflows**: For complex requests spanning multiple phases, sequence agents appropriately
7. **Execute and Document**: Route to selected agent(s), update Canvas, preserve context
### Multi-Agent Coordination
For requests requiring multiple agents, coordinate in sequence:
**Example: Initial Business Model**
1. market-intelligence → Define TAM/SAM/SOM, segments (Canvas 01-04, 06)
2. problem-solution-fit → Validate problem-solution fit (Canvas 05, 09)
3. value-proposition → Define unique value and positioning (Canvas 07-08)
4. business-model → Design pricing and revenue streams (Canvas 11-13)
5. validation → List key assumptions to test (Canvas 10)
**Example: Pre-Launch Preparation**
1. validation → Validate remaining assumptions (Canvas 10)
2. business-model → Finalize pricing (Canvas 11)
3. go-to-market → Create launch playbook (Canvas 14-15)
4. execution → Build launch implementation plan (artifacts/marketing/)
**Example: Post-Discovery Pivot**
1. market-intelligence → Re-analyze market with new insights (Canvas 01-06)
2. problem-solution-fit → Revalidate problem-solution fit (Canvas 05, 09)
3. validation → Design pivot validation experiments (Canvas 10)
4. execution → Plan pivot implementation
---
## Canvas Integration
### Canvas as Startup Playbook Output
**Path:** `strategy/canvas/`
This skill's primary mission is to populate all 16 Canvas sections with validated information:
```
strategy/canvas/
├── 00-business-model-mode.md ← foundation-builder (Phase 0)
├── 01-context.md ← market-intelligence
├── 02-constraints.md ← market-intelligence
├── 03-opportunity.md ← market-intelligence
├── 04-segments.md ← market-intelligence
├── 05-problem.md ← problem-solution-fit
├── 06-competitive.md ← market-intelligence
├── 07-uvp.md ← value-proposition
├── 08-advantage.md ← value-proposition
├── 09-solution.md ← problem-solution-fit
├── 10-assumptions.md ← validation (all agents contribute)
├── 11-pricing.md ← business-model
├── 12-costs.md ← business-model
├── 13-metrics.md ← business-model
├── 14-growth.md ← go-to-market
└── 15-gtm.md ← go-to-market
```
**Mode-Aware Note:** The 00-business-model-mode.md file determines which market research skill is used and how decisions are prioritized throughout Canvas building.
### Assumptions Tracking (Canvas 10)
All agents contribute hypotheses to `10-assumptions.md` during pre-launch:
```markdown
### A1: Target Market Size
**Status:** 📝 DRAFT
**Hypothesis:** AI-powered fashion recommendations TAM is $2B
**Confidence:** 40%
**Source:** market-intelligence agent, Gartner reports
**Test:** Pre-launch survey with 100 target customers
**Validation Criteria:** >60% express strong interest
```
**Status progression:**
- 📝 DRAFT → Pre-validation hypothesis
- ⏳ TESTING → Validation experiment running
- ✅ VALIDATED → Experiment confirmed hypothesis
- ❌ INVALIDATED → Experiment rejected hypothesis, pivot needed
### Launch Completion
**When Canvas is complete + validated:**
1. All 15 sections populated
2. Key assumptions (10-assumptions.md) tested
3. First customers acquired
4. Revenue > $0
**Then:**
- Canvas becomes foundation for operational decisions
- Begin managing business through evidence-based operations
---
### Context Management
To maintain effectiveness within subscription constraints:
1. **Documentation-First**: Every agent interaction produces Canvas updates or artifacts
2. **Git Commits**: Frequent commits preserve implementation context
3. **Session Planning**: Plan agent usage to stay within rate limits (~30-45 Claude messages per 3-4 hours)
4. **Task Batching**: Group related tasks to minimize context switching
5. **Canvas-Centric**: All strategic decisions flow through Canvas updates
---
## Agent Sub-Skills
Each of the 10 agents is implemented as a sub-skill within this orchestrator:
### Core Agents
- `market-intelligence/` - Market analysis and competitive intelligence
- `problem-solution-fit/` - Problem validation and solution design
- `value-proposition/` - Value articulation and positioning
- `business-model/` - Business model design and unit economics
- `validation/` - Hypothesis testing and experiment design
- `go-to-market/` - GTM strategy and launch planning
### Specialist Agents
- `funding/` - Fundraising strategy and investor relations
- `regulatory/` - Compliance and regulatory navigation
- `retention-optimizer/` - Post-first-customers retention analysis
**Note**: Sub-skills are invoked by this orchestrator based on the routing logic above. Each sub-skill contains specialized prompts, frameworks, and templates for its domain.
---
## Success Metrics
Track these metrics to ensure pre-launch progress:
**Canvas Completion**: 16/16 sections populated with validated info (including 00-business-model-mode.md)
**Assumption Validation**: >70% of key assumptions validated before launch
**MVP Velocity**: 1.5-2x solo developer productivity
**Cost Efficiency**:
- VENTURE: Burn rate tracked, runway >12 months
- BOOTSTRAP: <$500/month burn, profitable within 3 months
**Launch Readiness**: All go-to-market elements defined and ready
**Mode-Specific Success:**
- **VENTURE:** TAM validated >$1B, growth path to $100M ARR clear
- **BOOTSTRAP:** First customer within 4 weeks, path to $5k MRR defined
---
## Usage Examples
### Example 0: Business Model Mode Selection
**User Request**: "I'm starting a B2B SaaS, should I use VENTURE or BOOTSTRAP mode?"
**Orchestrator Action**:
1. Detect phase: Mode Selection (0)
2. Ask clarifying questions:
- Are you funded or planning to raise?
- What's your timeline to profitability?
- Is this a billion-dollar market or profitable niche?
3. Create: `strategy/canvas/00-business-model-mode.md`
4. Expected output: Mode declaration (VENTURE/BOOTSTRAP/HYBRID) with rationale
**Example Response for Bootstrap:**
```markdown
**Active Mode:** BOOTSTRAP
**Rationale:**
- Self-funded, need profitability within 3 months
- B2B SaaS with clear monetization path
- Targeting profitable niche ($50M market)
- Exit optional, cash flow primary goal
```
### Example 1: Initial Market Discovery (BOOTSTRAP Mode)
**User Request**: "Help me understand the competitive landscape for AI-powered beauty recommendations"
**Orchestrator Action**:
1. Check mode: Read `00-business-model-mode.md` → BOOTSTRAP
2. Detect phase: Discovery (0a)
3. Route to: market-intelligence agent + **market-research-bootstrap skill**
4. Canvas targets: 01-context, 02-constraints, 03-opportunity, 04-segments, 06-competitive
5. Expected output:
- Current spend mapping (who pays what today)
- Budget holder identification
- Arbitrage opportunities
- Competitive pricing analysis
- Q1 revenue potential
### Example 1b: Initial Market Discovery (VENTURE Mode)
**User Request**: "Help me understand the competitive landscape for AI-powered beauty recommendations"
**Orchestrator Action**:
1. Check mode: Read `00-business-model-mode.md` → VENTURE
2. Detect phase: Discovery (0a)
3. Route to: market-intelligence agent + **market-research-venture skill**
4. Canvas targets: 01-context, 02-constraints, 03-opportunity, 04-segments, 06-competitive
5. Expected output:
- TAM/SAM/SOM analysis ($XB market)
- Market growth rate (CAGR projections)
- Competitive positioning and moat opportunities
- 10x scale potential
- Defensibility assessment
### Example 2: Business Model Design
**User Request**: "How should we price our AI fashion recommendations service?"
**Orchestrator Action**:
1. Detect phase: Definition (0b)
2. Route sequence: business-model agent (primary) + value-proposition (supporting)
3. Canvas targets: 11-pricing, 12-costs, 13-metrics
4. Expected output: Pricing strategy with 3 tiers, unit economics, key metrics
### Example 3: MVP Validation
**User Request**: "We built the MVP, how do we validate it works?"
**Orchestrator Action**:
1. Detect phase: Validation (1a)
2. Route to: validation agent (primary) + execution agent (supporting)
3. Canvas target: 10-assumptions (update status)
4. Expected output: Validation experiment design, success criteria, 2-week test plan
### Example 4: Launch Preparation
**User Request**: "We're launching in 3 weeks, what do we need?"
**Orchestrator Action**:
1. Detect phase: Launch (1c)
2. Route sequence: go-to-market → execution
3. Canvas targets: 14-growth, 15-gtm
4. Outputs: Launch playbook, initial marketing materials in `artifacts/marketing/`
5. Expected output: 3-week sprint plan, launch checklist, completion criteria
---
## Best Practices
### For Effective Pre-Launch
1. **Start with Discovery**: Complete Canvas 01-06 before moving to Definition
2. **Validate Before Building**: Test assumptions (Canvas 10) before MVP development
3. **Document Everything**: All strategic decisions go into Canvas
4. **Iterate Quickly**: Use validation agent frequently to test hypotheses
5. **Complete Thoroughly**: Don't declare launch complete until Canvas is validated with real customers
### For Canvas Population
1. **One Section at a Time**: Focus agents on completing Canvas sections sequentially
2. **Evidence-Based**: Every Canvas claim needs source (customer interviews, data, research)
3. **Track Assumptions**: Every hypothesis goes into 10-assumptions.md with validation plan
4. **Update Status**: Move assumptions from DRAFT → TESTING → VALIDATED/INVALIDATED
5. **Maintain Quality**: Clean, complete Canvas is the foundation for post-launch operations
### For Budget Optimization
1. **Batch Canvas Updates**: Update multiple sections in one session
2. **Use Right Tool for Right Job**: Strategy in Claude, validation in GPT, boilerplate in Copilot
3. **Leverage Templates**: Reuse Canvas section templates
4. **Front-Load Discovery**: Thorough discovery (0a) prevents expensive pivots later
---
## Launch Completion Checklist
Before declaring pre-launch work complete:
- [ ] **Business model mode declared** (00-business-model-mode.md)
- [ ] All 16 Canvas sections populated
- [ ] Key assumptions (>70%) validated
- [ ] MVP launched to first customers
- [ ] Revenue > $0 (even if just $1)
- [ ] Customer feedback collected
- [ ] Initial metrics tracked (13-metrics.md)
- [ ] GTM channels identified and tested (15-gtm.md)
- [ ] Business foundation ready for operational scaling
**Mode-Specific Completion:**
**VENTURE Mode:**
- [ ] TAM validated >$1B
- [ ] Burn rate and runway tracked
- [ ] Growth path to $100M ARR defined
- [ ] Fundraising timeline (if applicable)
**BOOTSTRAP Mode:**
- [ ] First paying customer acquired
- [ ] Path to profitability within 3 months clear
- [ ] Positive cash flow or near-positive
- [ ] Unit economics profitable
**Pre-launch work is complete when:**
- Canvas is validated with real customer data
- Product-market fit signals are emerging
- Mode-appropriate success criteria met
- Ready to shift from building to scaling
---
## Remember
This skill is for **building the business from zero to launch**. Focus on:
- **FIRST:** Declare business model mode (VENTURE/BOOTSTRAP/HYBRID)
- Complete all 16 Canvas sections with validated data
- Use mode-appropriate market research (venture vs bootstrap skills)
- Test assumptions before scaling
- Achieve mode-appropriate traction:
- VENTURE: Validate TAM >$1B, define path to $100M ARR
- BOOTSTRAP: First customer, path to profitability within 3 months
- Build solid foundation for operational phase
**Critical:** The business model mode determines:
- Which market research skill to invoke
- How success is measured
- What metrics matter
- Decision prioritization throughout Canvas building
The startup-playbook mission is complete when you have customers, revenue, a validated Canvas, and mode-appropriate success criteria met. Don't rush to operations before the foundation is solid.