Getting Started

Quick Start

Get from zero to your first AI-generated PR in under 5 minutes.

1. Install Pivota

Clone the deployment repo, configure your environment, and start the platform:

# Pull and start Pivota
git clone https://github.com/pivota-dev/deploy.git
cd deploy
cp .env.example .env
# Edit .env with your API keys

docker compose up -d

# Verify
curl http://localhost:3000/health
# => {"status": "ok"}

2. Create Your First Workspace

Open the Pivota UI and create a workspace, or use the CLI:

# Open http://localhost:3000 in your browser
# Log in with default admin credentials
# Click "New Workspace" from the dashboard

# Or use the CLI:
/pivota_spec:new-project

? What do you want to build?
> A permit processing application

? Who is the target user?  
> Government agency staff processing permit applications

? Key constraints?
> Must integrate with existing PostgreSQL database
> FIPS compliance required
> Mobile-responsive UI

3. Generate Spec Documents

Pivota captures your requirements and generates a full spec document suite:

# Pivota generates:
✓ PROJECT.md — Project definition
✓ REQUIREMENTS.md — Captured requirements  
✓ ROADMAP.md — Phased execution plan
✓ STATE.md — Current project state
✓ research/ — Domain research artifacts

# Generate full spec documents:
/pivota_spec:generate-spec-docs

✓ PRD — Product Requirements Document
✓ Personas — User persona definitions
✓ FRD — Functional Requirements Document
✓ TechArch — Technical Architecture
✓ UserStories — User stories with acceptance criteria
✓ Journeys — User journey maps
✓ JTBD — Jobs to be Done analysis
✓ StoryMap — Epic/story mapping

4. Plan and Execute

Discuss the approach, plan the phase, then execute:

# Discuss implementation approach
/pivota_spec:discuss-phase 1

# Plan the phase (research + plan + verify)
/pivota_spec:plan-phase 1

✓ Researching domain patterns...
✓ Plan created: 6 tasks in 2 execution waves
✓ Plan verified against requirements

# Execute (or drag kanban card to "In Progress")
/pivota_spec:execute-phase 1

✓ Daytona sandbox provisioned
✓ OpenCode agent launched
✓ Task 1/6: Database schema — COMPLETE
✓ Task 2/6: API endpoints — COMPLETE  
✓ Task 3/6: Frontend components — COMPLETE
✓ Task 4/6: Integration tests — COMPLETE
✓ Task 5/6: E2E tests — COMPLETE
✓ Task 6/6: Documentation — COMPLETE

✓ All tests passing (24/24)
✓ PR #1 created: feature/permit-api

5. Verify

Verify the work against requirements and run a full milestone audit:

# Verify against requirements
/pivota_spec:verify-work 1

✓ Requirements coverage: 100%
✓ Tests: 24/24 passing
✓ Evidence pack generated
✓ Traceability matrix updated
✓ Phase 1: COMPLETE

# Full milestone audit
/pivota_spec:audit-milestone

✓ Definition of done: MET
✓ All phases verified
✓ Evidence packs: 6 generated
✓ Traceability: 100% coverage

Next Steps