Imported from kivo360/OmoiOS (.claude/skills/spec-driven-dev/SKILL.md). Install upstream with npx skills add kivo360/OmoiOS --skill spec-driven-dev. Copyright stays with the author.

Spec-Driven Development

A systematic workflow for turning feature ideas into actionable work items that AI agents can execute.

Architecture Note: This skill is designed to work with a state machine orchestrator that executes each phase as a separate Claude SDK session. Each phase saves checkpoints to the database and can resume from any point after failure. See docs/spec-execution-stability.md for the full architecture.

🔴 CRITICAL: Read This Skill Document Thoroughly

YOU MUST follow this skill document exactly. This is not optional guidance—it is the required workflow.

Before Creating ANY Spec Files:

1. READ THIS ENTIRE SKILL DOCUMENT - Don't skim. Read every section to understand the required formats.
2. CHECK EXISTING FILES - Run ls -la .omoi_os/ to see what already exists. Don't duplicate.
3. REFERENCE THE TEMPLATES - Every file type has a specific format. Copy the exact structure.
4. USE THE CLI TOOLS - Validate with python spec_cli.py validate before syncing.

During Spec Creation:

1. REFER BACK TO THIS DOCUMENT OFTEN - When unsure about format, re-read the relevant section.
2. COPY FRONTMATTER EXACTLY - Don't improvise. Use the exact field names shown in templates.
3. CHECK YOUR WORK - After creating files, run validation to catch errors early.

Output Requirements:

• ALL files MUST have YAML frontmatter - No exceptions
• ALL frontmatter fields MUST match the templates - Use exact field names
• ALL specs MUST be synced - Run python spec_cli.py sync push when done
• ALL IDs MUST follow conventions - TKT-001, TSK-001, REQ-FEATURE-001, etc.

If You're Unsure:

1. Re-read this skill document - The answer is here
2. Look at the Concrete Example section - Full file contents are provided
3. Run validation - python spec_cli.py validate will tell you what's wrong

🔄 State Machine Architecture

Overview

Spec generation runs as a state machine with discrete phases. Each phase:

1. Receives context from previous phases
2. Executes a focused, time-boxed Claude SDK session
3. Validates output with evaluators
4. Saves checkpoint to database and file system
5. Stores session transcript for potential resumption

┌─────────────┐    ┌─────────────┐    ┌─────────────┐    ┌─────────────┐    ┌─────────────┐
│   EXPLORE   │───▶│ REQUIREMENTS│───▶│   DESIGN    │───▶│   TASKS     │───▶│    SYNC     │
│  (codebase) │    │  (EARS fmt) │    │ (arch+data) │    │  (atomic)   │    │  (to API)   │
└─────────────┘    └─────────────┘    └─────────────┘    └─────────────┘    └─────────────┘
       │                  │                  │                  │                  │
       ▼                  ▼                  ▼                  ▼                  ▼
   CHECKPOINT         CHECKPOINT         CHECKPOINT         CHECKPOINT         COMPLETE

Phase Definitions

Phase Context Flow

Each phase receives accumulated context from all previous phases:

# REQUIREMENTS phase receives:
{
    "exploration_context": {
        "project_type": "Next.js + FastAPI",
        "existing_models": [...],
        "conventions": {...},
        "related_to_feature": [...]
    },
    "feature_request": "User's original request"
}

# DESIGN phase receives:
{
    "exploration_context": {...},
    "requirements": [...],  # From REQUIREMENTS phase
    "feature_name": "feature-name"
}

# TASKS phase receives:
{
    "exploration_context": {...},
    "requirements": [...],
    "design": {...},  # From DESIGN phase
    "feature_name": "feature-name"
}

Evaluator Criteria

Each phase output is validated before proceeding:

Retry Logic

If validation fails:

1. Evaluator returns failure reasons
2. State machine retries the phase (max 3 attempts)
3. Retry prompt includes previous attempt and failure reasons
4. If all retries fail, phase is marked as failed with error details

Session Transcript Persistence

Session transcripts are stored for cross-sandbox resumption:

.omoi_os/
├── phase_data/
│   ├── explore.json           # EXPLORE phase output
│   ├── requirements.json      # REQUIREMENTS phase output
│   ├── design.json            # DESIGN phase output
│   └── tasks.json             # TASKS phase output
├── session_transcripts/
│   ├── explore.jsonl          # EXPLORE session transcript
│   ├── requirements.jsonl     # REQUIREMENTS session transcript
│   └── ...
└── checkpoints/
    └── state.json             # Current state checkpoint

Environment Variables

When running in a sandbox, these environment variables control execution:

🚨 MANDATORY: YAML Frontmatter on ALL Files

EVERY file you create in .omoi_os/ MUST begin with YAML frontmatter. This is NON-NEGOTIABLE.

Why Frontmatter is Required

1. Programmatic Parsing: The CLI tools, API sync, and orchestrator all parse frontmatter to understand file structure
2. Traceability: Frontmatter enables linking requirements → designs → tickets → tasks
3. Status Tracking: Status, priority, and dependencies are tracked via frontmatter fields
4. API Integration: When syncing to backend, frontmatter provides the structured data

Required Frontmatter by File Type

PRDs (.omoi_os/docs/prd-*.md):

---
id: PRD-{FEATURE}-001
title: {Feature Name} PRD
feature: {feature-name}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
status: draft  # draft | review | approved
author: Claude
---

Requirements (.omoi_os/requirements/*.md):

---
id: REQ-{FEATURE}-001
title: {Feature Name} Requirements
feature: {feature-name}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
status: draft  # draft | review | approved
category: functional  # functional | non-functional | constraint
priority: HIGH  # CRITICAL | HIGH | MEDIUM | LOW
prd_ref: docs/prd-{feature-name}.md
design_ref: designs/{feature-name}.md
---

Designs (.omoi_os/designs/*.md):

---
id: DESIGN-{FEATURE}-001
title: {Feature Name} Design
feature: {feature-name}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
status: draft  # draft | review | approved
requirements:
  - REQ-{FEATURE}-001
# API endpoints (synced to Design panel)
api_endpoints:
  - method: POST
    path: /api/v1/{resource}
    description: Create a new resource
    auth_required: true
    request_body: '{"field": "value"}'
    response: '{"id": "uuid", "field": "value"}'
  - method: GET
    path: /api/v1/{resource}/{id}
    description: Get resource by ID
    auth_required: true
    path_params: [id]
# Data models (synced to Design panel)
data_models:
  - name: ResourceModel
    description: Main resource entity
    table_name: resources
    typed_fields:
      - name: id
        type: uuid
        description: Unique identifier
        constraints: [primary_key]
      - name: name
        type: string
        description: Resource name
      - name: created_at
        type: timestamp
        default: now()
    relationships:
      - belongs_to User
      - has_many Items
---

Tickets (.omoi_os/tickets/TKT-*.md):

---
id: TKT-{NNN}
title: {Ticket Title}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
status: backlog  # backlog | analyzing | building | testing | done | blocked
priority: HIGH  # CRITICAL | HIGH | MEDIUM | LOW
estimate: M  # S | M | L | XL
design_ref: designs/{feature-name}.md
requirements:
  - REQ-{FEATURE}-FUNC-001
dependencies:
  blocked_by: []
  blocks: []
---

Tasks (.omoi_os/tasks/TSK-*.md):

---
id: TSK-{NNN}
title: {Task Title}
created: {YYYY-MM-DD}
status: pending  # pending | in_progress | review | done | blocked
parent_ticket: TKT-{NNN}
estimate: S  # S | M | L
type: implementation  # implementation | refactor | test | documentation | research | bugfix
dependencies:
  depends_on: []
  blocks: []
---

❌ Files Without Frontmatter Will Fail

Files missing frontmatter will:

• Not appear in spec_cli.py show commands
• Not sync to the API
• Not be tracked in traceability reports
• Not be picked up by the orchestrator

✅ Always Start With Frontmatter

When creating ANY file in .omoi_os/, your FIRST action should be writing the YAML frontmatter block, THEN the content.

🚨 MANDATORY: Use Spec CLI to Sync to Server

After creating files in .omoi_os/, you MUST sync them to the OmoiOS server using the spec CLI. This is NON-NEGOTIABLE.

Why Sync is Required

1. Visibility: Specs/tickets/tasks only appear in the dashboard after syncing
2. Orchestration: The orchestrator only picks up tasks from the database, not from files
3. Traceability: Server-side tracking enables dependency management and status updates
4. Persistence: Local files can be lost; server data persists across sandbox restarts

Spec CLI Location

The spec CLI is located at:

/root/.claude/skills/spec-driven-dev/scripts/spec_cli.py

Required Workflow: Create → Validate → Sync

EVERY time you create or modify files in .omoi_os/, follow this workflow:

# Step 1: Navigate to the spec CLI scripts directory
cd /root/.claude/skills/spec-driven-dev/scripts

# Step 2: Validate your specs (check for errors BEFORE syncing)
python spec_cli.py validate

# Step 3: Preview what will be synced (dry run)
# Note: OMOIOS_PROJECT_ID and OMOIOS_API_URL are auto-injected - no need to specify!
python spec_cli.py sync-specs diff

# Step 4: Sync requirements and designs to create specs
python spec_cli.py sync-specs push

# Step 5: Preview ticket/task sync
python spec_cli.py sync diff

# Step 6: Sync tickets and tasks
python spec_cli.py sync push

# Step 7: Verify traceability
python spec_cli.py api-trace

Environment Variables (Auto-Injected by Sandbox)

When the sandbox is created, the orchestrator automatically injects environment variables so that CLI tools and scripts work without manual configuration. This design means:

1. No manual URL configuration - The API client knows where to connect
2. No credential passing - Authentication is pre-configured
3. No project ID lookup - The context is already set

Auto-injected variables:

Fallback behavior: If running outside a sandbox (e.g., local development), the CLI falls back to https://api.omoios.dev as the default API URL. You can override any value by passing explicit arguments like --api-url or --project-id if needed.

Bottom line: Just run the commands - no configuration required inside sandboxes.

Quick Reference Commands

# Show all local specs
python spec_cli.py show all

# Show only tickets
python spec_cli.py show tickets

# Show only tasks with blocking status
python spec_cli.py show tasks

# Show dependency graph
python spec_cli.py show graph

# Show ready tasks (not blocked)
python spec_cli.py show ready

# List projects from server (uses OMOIOS_API_URL automatically)
python spec_cli.py projects

# View project details (uses OMOIOS_PROJECT_ID and OMOIOS_API_URL automatically)
python spec_cli.py project

❌ DO NOT Skip Syncing

If you only create local files without syncing:

• Tasks won't be picked up by agents
• Tickets won't appear in the dashboard
• Dependencies won't be tracked
• Status updates won't propagate

The Core Flow

The workflow maps to the state machine phases:

┌─────────────────────────────────────────────────────────────────────────────┐
│                        STATE MACHINE PHASES                                  │
│                                                                             │
│   EXPLORE ──────▶ REQUIREMENTS ──────▶ DESIGN ──────▶ TASKS ──────▶ SYNC   │
│     │                  │                 │              │            │      │
│   Analyze          Define WHAT       Define HOW     Break into    Push to  │
│   codebase         must happen       to build it    work items    API      │
│   context                                                                    │
│                                                                             │
│   explore.json   requirements/*.md  designs/*.md   tickets/*.md   API      │
│                                                    tasks/*.md     synced   │
└─────────────────────────────────────────────────────────────────────────────┘

Output Directory: All artifacts go in .omoi_os/ Sync Tool: Use spec_cli.py to push to the API State Machine: Each phase is a separate SDK session with checkpointing

Phase 1: EXPLORE (Most Important!)

State Machine Phase: EXPLORE - Timeout: 3 min, Max Turns: 25 Output: .omoi_os/phase_data/explore.json

Why This Phase Matters

The quality of everything downstream depends on deeply understanding the problem AND the existing codebase first. Never skip exploration. Rushing to create specs without understanding leads to wasted work and specs that don't align with existing patterns.

Step 1.1: Explore Existing Context

BEFORE asking questions, gather context:

# Check for existing documentation
Read docs/CLAUDE.md
Read docs/architecture/
ls .omoi_os/

# Search for related code
Grep for related service names, models, patterns
Read existing implementations this feature will integrate with

# Check for prior work
ls .omoi_os/requirements/
ls .omoi_os/tickets/
git log --oneline -20

Step 1.2: Ask Discovery Questions (5-15 Questions)

Structure your questions in categories:

Problem & Value (2-3 questions)

• What specific problem does this solve? What pain exists today?
• What happens if we DON'T build this?
• How will we measure success? What metrics matter?

Users & Journeys (2-3 questions)

• Who are the primary users? Secondary?
• What's the happy path user journey?
• What are the edge cases and error scenarios?

Scope & Boundaries (2-3 questions)

• What is explicitly IN scope?
• What is explicitly OUT of scope? (Very important!)
• Are there existing features this overlaps with?

Technical Context (3-5 questions)

• What existing systems/services will this integrate with?
• What data does this need? Where does it come from?
• Are there performance requirements (latency, throughput, scale)?
• What security/privacy considerations apply?
• Are there any hard technical constraints?

Trade-offs & Risks (2-3 questions)

• Are there multiple valid approaches? Which should we explore?
• What are the trade-offs between approaches?
• What could go wrong? What are the risks?
• What's the timeline/priority?

Step 1.3: Summarize Understanding

After questions are answered, write a summary:

## Feature Summary

**Name**: feature-name (kebab-case)
**One-liner**: Brief description of what this does

**Problem Statement**:
[2-3 sentences about the pain point this solves]

**User Stories**:
1. As a [user], I can [action] so that [benefit]
2. As a [user], I can [action] so that [benefit]
3. ...

**Scope**:
- IN: [list what's included]
- OUT: [list what's explicitly excluded]

**Technical Constraints**:
- [constraint 1]
- [constraint 2]

**Risks Identified**:
- [risk 1]
- [risk 2]

**Success Metrics**:
- [metric 1]
- [metric 2]

Get user confirmation before proceeding!

Step 1.4: Save Exploration Output (State Machine)

When running in the state machine, save the exploration context as JSON:

# .omoi_os/phase_data/explore.json
{
    "project_type": "Next.js 15 + FastAPI backend",
    "structure": {
        "frontend": "frontend/src/",
        "backend": "backend/omoi_os/",
        "api_routes": "backend/omoi_os/api/routes/",
        "models": "backend/omoi_os/models/",
        "services": "backend/omoi_os/services/"
    },
    "existing_models": [
        {"name": "Spec", "file": "backend/omoi_os/models/spec.py", "fields": ["id", "title", "description"]},
        {"name": "Ticket", "file": "backend/omoi_os/models/ticket.py", "fields": ["id", "title", "status"]}
    ],
    "conventions": {
        "naming": "snake_case for backend, camelCase for frontend",
        "testing": "pytest for backend, vitest for frontend",
        "patterns": ["Repository pattern", "Service layer", "Pydantic schemas"]
    },
    "related_to_feature": [
        {"name": "EventBusService", "file": "services/event_bus.py", "relevance": "Publish events for webhooks"},
        {"name": "TaskQueue", "file": "services/task_queue.py", "relevance": "Background processing"}
    ],
    "feature_summary": {
        "name": "webhook-notifications",
        "problem": "External systems cannot subscribe to OmoiOS events",
        "scope_in": ["Webhook subscriptions", "HMAC signing", "Retry logic"],
        "scope_out": ["UI management", "Rate limiting (future)"]
    }
}

Evaluator Criteria:

• ✓ Has project_type (non-empty string)
• ✓ Has structure (object with at least one key)
• ✓ Has conventions (object with naming patterns)
• ✓ Has related_to_feature (list, can be empty for greenfield)

Phase 2: PRD (Product Requirements Document)

Purpose

The PRD captures the "why" and "what" at a high level. It's the vision document that everything else traces back to.

Location

.omoi_os/docs/prd-{feature-name}.md

Template

---
id: PRD-{FEATURE}-001
title: {Feature Name} PRD
feature: {feature-name}
created: {date}
updated: {date}
status: draft
author: Claude
---

# {Feature Name}

## Executive Summary

[2-3 paragraph overview of the feature, the problem it solves, and why it matters]

## Problem Statement

### Current State
[Describe the pain point that exists today]

### Desired State
[Describe what the world looks like after this feature ships]

### Impact of Not Building
[What happens if we don't build this?]

## Goals & Success Metrics

### Primary Goals
1. [Goal 1]
2. [Goal 2]

### Success Metrics
| Metric | Current | Target | How Measured |
|--------|---------|--------|--------------|
| [Metric 1] | [baseline] | [target] | [method] |

## User Stories

### Primary User: {User Type}

1. **{Story Title}**
   As a {user}, I want to {action} so that {benefit}.

   Acceptance Criteria:
   - [ ] {criterion 1}
   - [ ] {criterion 2}

2. ...

## Scope

### In Scope
- [Feature/capability 1]
- [Feature/capability 2]

### Out of Scope
- [Explicitly excluded 1]
- [Explicitly excluded 2]

### Future Considerations
- [Things we might add later]

## Constraints

### Technical Constraints
- [Constraint 1]
- [Constraint 2]

### Business Constraints
- [Timeline, budget, etc.]

## Risks & Mitigations

| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| [Risk 1] | High/Med/Low | High/Med/Low | [How to mitigate] |

## Dependencies

- [Dependency 1 - what we need from other teams/systems]
- [Dependency 2]

## Open Questions

- [ ] [Question that still needs answering]
- [ ] [Another question]

Phase 3: REQUIREMENTS

State Machine Phase: REQUIREMENTS (Timeout: 5 min / 20 turns) Transition: EXPLORE → REQUIREMENTS (requires ExplorationEvaluator pass) Next Phase: DESIGN (requires RequirementEvaluator pass)

Purpose

Requirements define the specific, testable behaviors the system must have. Use EARS format (Easy Approach to Requirements Syntax).

Location

.omoi_os/requirements/{feature-name}.md

EARS Format

WHEN [trigger/condition], THE SYSTEM SHALL [action/behavior].

Examples:

• WHEN a user submits valid credentials, THE SYSTEM SHALL create an authenticated session within 2 seconds.
• WHEN a notification is triggered, THE SYSTEM SHALL deliver it to the user within 5 seconds.
• WHEN an API rate limit is exceeded, THE SYSTEM SHALL return a 429 status with retry-after header.

Template

---
id: REQ-{FEATURE}-001
title: {Feature Name} Requirements
feature: {feature-name}
created: {date}
updated: {date}
status: draft
category: functional
priority: HIGH
prd_ref: docs/prd-{feature-name}.md
design_ref: designs/{feature-name}.md
---

# {Feature Name} Requirements

## Overview

[Brief description of what these requirements cover]

## Functional Requirements

### REQ-{FEATURE}-FUNC-001: {Requirement Title}

**Priority**: HIGH | MEDIUM | LOW
**Category**: Functional

WHEN {trigger condition}, THE SYSTEM SHALL {required behavior}.

**Acceptance Criteria**:
- [ ] {Testable criterion 1}
- [ ] {Testable criterion 2}

**Notes**: {Any additional context}

---

### REQ-{FEATURE}-FUNC-002: {Requirement Title}

...

## Non-Functional Requirements

### REQ-{FEATURE}-PERF-001: {Performance Requirement}

**Priority**: HIGH
**Category**: Performance

THE SYSTEM SHALL {performance requirement with measurable target}.

**Metrics**:
- P50 latency: < X ms
- P99 latency: < Y ms
- Throughput: > Z requests/second

---

### REQ-{FEATURE}-SEC-001: {Security Requirement}

**Priority**: HIGH
**Category**: Security

THE SYSTEM SHALL {security requirement}.

## Traceability

| Requirement ID | PRD Section | Design Section | Ticket |
|----------------|-------------|----------------|--------|
| REQ-{FEATURE}-FUNC-001 | User Stories #1 | Component A | TKT-001 |

Step 3.1: Save Requirements Output (State Machine)

After completing requirements, save to .omoi_os/phase_data/requirements.json:

{
    "requirements": [
        {
            "id": "REQ-{FEATURE}-FUNC-001",
            "category": "functional",
            "priority": "HIGH",
            "condition": "WHEN user submits form",
            "action": "THE SYSTEM SHALL validate and persist data",
            "acceptance_criteria": ["Criterion 1", "Criterion 2"]
        }
    ],
    "total_count": 15,
    "categories": {
        "functional": 10,
        "performance": 3,
        "security": 2
    }
}

RequirementEvaluator Criteria

Phase 4: DESIGN

State Machine Phase: DESIGN (Timeout: 5 min / 25 turns) Transition: REQUIREMENTS → DESIGN (requires RequirementEvaluator pass) Next Phase: TASKS (requires DesignEvaluator pass)

Purpose

The design document defines HOW to implement the requirements. It covers architecture, data models, APIs, and implementation details.

Location

.omoi_os/designs/{feature-name}.md

Template

---
id: DESIGN-{FEATURE}-001
title: {Feature Name} Design
feature: {feature-name}
created: {date}
updated: {date}
status: draft
requirements:
  - REQ-{FEATURE}-001
# API Endpoints - These sync to the Design panel's "API Specification" section
api_endpoints:
  - method: POST
    path: /api/v1/{resource}
    description: Create a new resource
    auth_required: true
    request_body: '{"name": "string", "config": {}}'
    response: '{"id": "uuid", "name": "string", "created_at": "timestamp"}'
    error_responses:
      400: Validation error
      401: Unauthorized
      409: Resource already exists
  - method: GET
    path: /api/v1/{resource}/{id}
    description: Get resource by ID
    auth_required: true
    path_params: [id]
    response: '{"id": "uuid", "name": "string", "status": "string"}'
    error_responses:
      404: Resource not found
  - method: GET
    path: /api/v1/{resource}
    description: List all resources
    auth_required: true
    query_params:
      limit: Maximum number of results
      offset: Pagination offset
      status: Filter by status
  - method: PUT
    path: /api/v1/{resource}/{id}
    description: Update resource
    auth_required: true
    path_params: [id]
  - method: DELETE
    path: /api/v1/{resource}/{id}
    description: Delete resource
    auth_required: true
    path_params: [id]
# Data Models - These sync to the Design panel's "Data Model" section
data_models:
  - name: {ResourceName}
    description: Main entity for this feature
    table_name: {table_name}
    typed_fields:
      - name: id
        type: uuid
        description: Unique identifier
        constraints: [primary_key]
      - name: name
        type: string
        description: Display name
        nullable: false
      - name: status
        type: string
        description: Current status
        default: "'pending'"
      - name: config
        type: jsonb
        description: Configuration data
        nullable: true
      - name: created_at
        type: timestamp
        default: now()
      - name: updated_at
        type: timestamp
        default: now()
    relationships:
      - belongs_to Project (project_id)
      - has_many Tasks
---

# {Feature Name} Design

## Overview

[Brief description of the technical approach]

## Architecture Overview

```mermaid
flowchart TB
    subgraph External
        User[User]
        ExtAPI[External API]
    end

    subgraph System
        API[API Layer]
        Service[Service Layer]
        DB[(Database)]
    end

    User --> API
    API --> Service
    Service --> DB
    Service --> ExtAPI

Component Diagram

[Describe the main components and their responsibilities]

Data Model

The data models are defined in the frontmatter above and will sync to the Design panel. Additional SQL details can be added here:

Database Schema

CREATE TABLE {table_name} (
    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
    {column} {type} {constraints},
    created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW(),
    updated_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
);

CREATE INDEX idx_{table}_{column} ON {table}({column});

Pydantic Models

from pydantic import BaseModel, Field
from datetime import datetime
from uuid import UUID

class {ModelName}(BaseModel):
    id: UUID
    {field}: {type} = Field(description="{description}")
    created_at: datetime
    updated_at: datetime

API Specification

The API endpoints are defined in the frontmatter above and will sync to the Design panel. Additional request/response examples can be documented here:

Request/Response Examples

POST /api/v1/{resource}

Request:

{
  "name": "My Resource",
  "config": {}
}

Response (201):

{
  "id": "550e8400-e29b-41d4-a716-446655440000",
  "name": "My Resource",
  "status": "pending",
  "created_at": "2025-01-01T00:00:00Z"
}

Error Response (400):

{
  "error": "validation_error",
  "message": "Field 'name' is required"
}

Implementation Details

Algorithm/Logic

[Describe any complex algorithms or business logic]

# Pseudocode for complex logic
def process_feature(input):
    # Step 1: Validate
    # Step 2: Transform
    # Step 3: Persist
    # Step 4: Notify
    pass

State Machine (if applicable)

stateDiagram-v2
    [*] --> Created
    Created --> Processing : start
    Processing --> Completed : success
    Processing --> Failed : error
    Failed --> Processing : retry
    Completed --> [*]

Integration Points

Configuration

Testing Strategy

Unit Tests

• {Component} - {What to test}

Integration Tests

• {Integration point} - {What to test}

E2E Tests

• {User journey} - {What to test}

Security Considerations

• {Security consideration 1}
• {Security consideration 2}

Performance Considerations

• {Performance consideration 1}
• {Performance consideration 2}

Open Questions

• {Technical question that needs resolution}

### Step 4.1: Save Design Output (State Machine)

After completing design, save to `.omoi_os/phase_data/design.json`:

```json
{
    "api_endpoints": [
        {
            "method": "POST",
            "path": "/api/v1/resource",
            "description": "Create resource",
            "request_schema": "ResourceCreate",
            "response_schema": "Resource"
        }
    ],
    "data_models": [
        {
            "name": "Resource",
            "table_name": "resources",
            "fields": [
                {"name": "id", "type": "uuid", "primary_key": true},
                {"name": "name", "type": "string", "nullable": false}
            ]
        }
    ],
    "components": [
        {
            "name": "ResourceService",
            "layer": "service",
            "responsibilities": ["CRUD operations", "Validation"]
        }
    ]
}

DesignEvaluator Criteria

Phase 5: TICKETS (Combined with Tasks in State Machine)

Note: In the state machine, TICKETS and TASKS are combined into a single TASKS phase for generation efficiency. The evaluator validates both ticket and task structures together.

Purpose

Tickets are the parent work items that group related tasks. One ticket typically maps to one major component or capability from the design.

Location

.omoi_os/tickets/TKT-{NNN}.md

ID Convention

Generate sequential IDs: TKT-001, TKT-002, etc.

Check existing: ls .omoi_os/tickets/

Template

---
id: TKT-{NNN}
title: {Ticket Title}
created: {YYYY-MM-DD}
updated: {YYYY-MM-DD}
status: backlog
priority: HIGH
estimate: M
design_ref: designs/{feature-name}.md
requirements:
  - REQ-{FEATURE}-FUNC-001
  - REQ-{FEATURE}-FUNC-002
dependencies:
  blocked_by: []
  blocks: []
---

# TKT-{NNN}: {Ticket Title}

## Description

[Clear description of what this ticket accomplishes]

## Acceptance Criteria

- [ ] {Criterion 1 - specific and testable}
- [ ] {Criterion 2}
- [ ] {Criterion 3}
- [ ] All tests pass
- [ ] Code reviewed and approved

## Technical Notes

[Any technical context or decisions relevant to implementation]

## Tasks

- TSK-{NNN}: {Task 1 title}
- TSK-{NNN}: {Task 2 title}
- TSK-{NNN}: {Task 3 title}

## Dependencies

**Blocked By**: {List tickets that must complete first, or "None"}
**Blocks**: {List tickets waiting on this one, or "None"}

## Estimate

**Size**: S | M | L | XL
**Rationale**: {Why this estimate}

Ticket Generation Rules

1. One ticket per major component from the design
2. Clear acceptance criteria mapped from requirements
3. Dependencies explicitly listed - what blocks this, what this blocks
4. Reasonable scope - if > 5 tasks, consider splitting the ticket

Phase 6: TASKS

State Machine Phase: TASKS (Timeout: 3 min / 15 turns) Transition: DESIGN → TASKS (requires DesignEvaluator pass) Next Phase: SYNC (requires TaskEvaluator pass)

Purpose

Tasks are atomic units of work that an AI agent can complete in a single session. They should be specific, self-contained, and have clear deliverables.

Location

.omoi_os/tasks/TSK-{NNN}.md

ID Convention

Generate sequential IDs: TSK-001, TSK-002, etc.

Check existing: ls .omoi_os/tasks/

Template

---
id: TSK-{NNN}
title: {Task Title}
created: {YYYY-MM-DD}
status: pending
parent_ticket: TKT-{NNN}
estimate: S
type: implementation
dependencies:
  depends_on: []
  blocks: []
---

# TSK-{NNN}: {Task Title}

## Objective

[One sentence describing what this task accomplishes]

## Context

[Brief context about why this task exists and how it fits into the bigger picture]

## Deliverables

- [ ] `{path/to/file.py}` - {What this file does}
- [ ] `{path/to/test_file.py}` - {Tests for the above}

## Implementation Notes

[Specific guidance for how to implement this]

```python
# Example code structure or pseudocode
class FeatureService:
    def do_thing(self):
        pass

Acceptance Criteria

• {Specific, testable criterion}
• {Another criterion}
• All new code has tests
• Tests pass

Dependencies

Depends On: {List task IDs that must complete first, or "None"} Blocks: {List task IDs waiting on this one, or "None"}

Estimate

Size: S (< 2 hours) | M (2-4 hours) | L (4-8 hours)

### Task Generation Rules

1. **Atomic** - Can be completed in a single focused session (1-4 hours ideal)
2. **Self-contained** - All context needed is in the task
3. **Clear deliverable** - Specific files, tests, or outcomes
4. **Testable** - Clear acceptance criteria that can be verified
5. **Dependencies explicit** - What must complete before this can start

### Task Types

- `implementation` - Write new code
- `refactor` - Improve existing code
- `test` - Write tests
- `documentation` - Write docs
- `research` - Investigate options
- `bugfix` - Fix a specific bug

### Step 6.1: Save Tasks Output (State Machine)

After completing tasks, save to `.omoi_os/phase_data/tasks.json`:

```json
{
    "tickets": [
        {
            "id": "TKT-001",
            "title": "Implement Resource API",
            "priority": "HIGH",
            "estimate": "M",
            "tasks": ["TSK-001", "TSK-002", "TSK-003"],
            "dependencies": {"blocked_by": [], "blocks": ["TKT-002"]}
        }
    ],
    "tasks": [
        {
            "id": "TSK-001",
            "title": "Create ResourceService",
            "parent_ticket": "TKT-001",
            "type": "implementation",
            "estimate": "M",
            "deliverables": ["omoi_os/services/resource_service.py"],
            "dependencies": {"depends_on": [], "blocks": ["TSK-002"]}
        }
    ],
    "dependency_graph": {
        "TSK-001": [],
        "TSK-002": ["TSK-001"],
        "TSK-003": ["TSK-002"]
    }
}

TaskEvaluator Criteria

Phase 7: SYNC TO API

State Machine Phase: SYNC (Timeout: 2 min / 10 turns) Transition: TASKS → SYNC (requires TaskEvaluator pass) Final State: COMPLETE (spec ready for execution)

Using spec_cli.py

After creating all artifacts locally, sync them to the OmoiOS API:

# Navigate to scripts directory
cd /root/.claude/skills/spec-driven-dev/scripts

# 1. Validate local specs first
python spec_cli.py validate

# 2. Preview what you have
python spec_cli.py show all

# 3. Find your project ID (if not in env)
python spec_cli.py projects

# 4. Dry-run sync to see what would change
python spec_cli.py sync diff

# 5. Push to API
python spec_cli.py sync push

# 6. Verify in API
python spec_cli.py project

Sync Behavior

• CREATE: New tickets/tasks are created
• UPDATE: Existing items with changed descriptions are updated
• SKIP: Items already in sync are skipped

Complete Example Workflow

User Request

"I want to add webhook notifications so external systems can subscribe to events"

Step 1: Discover

Claude: Before designing, let me explore the codebase and ask some questions.

[Reads docs/CLAUDE.md, checks .omoi_os/ for existing specs, searches for event-related code]

Found: We have an EventBusService for internal pub/sub. No existing webhook specs.

Now I need to understand the feature better:

**Problem & Value:**
1. What specific external systems need to receive notifications?
2. What events should be subscribable? (task complete, PR ready, agent stuck?)
3. What happens if a webhook delivery fails?

**Scope:**
4. Should users configure webhooks via UI, API, or both?
5. Do we need authentication for webhook endpoints (HMAC signing)?
6. Is retry logic needed? How many retries?

**Technical:**
7. Should webhooks integrate with the existing EventBusService?
8. What's the expected volume of webhook deliveries per day?
9. Do we need rate limiting per subscriber?

**Constraints:**
10. Are there any specific protocols required (Slack, Discord, custom)?
11. What's the timeline for this feature?

Step 2: User Answers

"We need Slack and Discord initially, plus generic HTTP. Events: task_completed, ticket_status_changed, agent_error. HMAC signing required. 3 retries with exponential backoff. Use existing EventBusService. Volume: ~1000/day initially."

Step 3: Create PRD

Write .omoi_os/docs/prd-webhook-notifications.md

Step 4: Create Requirements

Write .omoi_os/requirements/webhook-notifications.md with EARS-format requirements

Step 5: Create Design

Write .omoi_os/designs/webhook-notifications.md with architecture, data models, APIs

Step 6: Create Tickets

• .omoi_os/tickets/TKT-001.md - Webhook Infrastructure (models, service, delivery)
• .omoi_os/tickets/TKT-002.md - Slack/Discord Formatters

Step 7: Create Tasks

For TKT-001:

• TSK-001.md - Add webhook URL fields to Organization model
• TSK-002.md - Create database migration
• TSK-003.md - Implement WebhookDeliveryService
• TSK-004.md - Implement WebhookNotificationService
• TSK-005.md - Update API routes for webhook management
• TSK-006.md - Write tests

For TKT-002:

• TSK-007.md - Implement SlackWebhookFormatter
• TSK-008.md - Implement DiscordWebhookFormatter

Step 8: Sync to API

cd /root/.claude/skills/spec-driven-dev/scripts
python spec_cli.py validate
python spec_cli.py sync push

Concrete Example: Full File Contents

This section shows exact file contents you should create. Copy these patterns for your specs.

Example Ticket File

File: .omoi_os/tickets/TKT-001.md

---
id: TKT-001
title: Webhook Infrastructure Setup
created: 2025-01-15
updated: 2025-01-15
status: backlog
priority: HIGH
type: feature
feature: webhook-notifications
estimate: 3d
requirements:
  - REQ-WEBHOOK-FUNC-001
  - REQ-WEBHOOK-FUNC-002
dependencies:
  blocked_by: []
  blocks:
    - TKT-002
---

# TKT-001: Webhook Infrastructure Setup

## Objective

Set up the core webhook infrastructure including database models, delivery service with retry logic, and API endpoints for webhook management.

## Scope

### In Scope
- WebhookSubscription database model
- WebhookDelivery model for tracking delivery attempts
- WebhookDeliveryService with exponential backoff retry
- CRUD API endpoints for webhook subscriptions
- HMAC signature generation for payload verification

### Out of Scope
- Platform-specific formatters (Slack, Discord) - covered in TKT-002
- UI for webhook management - future ticket

## Acceptance Criteria

- [ ] WebhookSubscription model with url, secret, events[], active fields
- [ ] WebhookDelivery model tracks each delivery attempt with status
- [ ] Service delivers webhooks with 3 retries and exponential backoff
- [ ] All endpoints require authentication
- [ ] Webhook payloads include HMAC-SHA256 signature header
- [ ] Unit tests cover delivery retry logic
- [ ] Integration tests verify end-to-end webhook delivery

## Technical Notes

- Integrate with existing EventBusService to subscribe to events
- Use background task for async delivery (don't block event processing)
- Store webhook secret encrypted at rest

## Tasks

- TSK-001: Add webhook models to database
- TSK-002: Create database migration
- TSK-003: Implement WebhookDeliveryService
- TSK-004: Implement WebhookNotificationService
- TSK-005: Add API routes for webhook CRUD
- TSK-006: Write comprehensive tests

Example Task File

File: .omoi_os/tasks/TSK-003.md

---
id: TSK-003
title: Implement WebhookDeliveryService
created: 2025-01-15
status: pending
priority: HIGH
type: implementation
parent_ticket: TKT-001
estimate: 4h
dependencies:
  depends_on:
    - TSK-001
    - TSK-002
  blocks:
    - TSK-004
---

# TSK-003: Implement WebhookDeliveryService

## Objective

Create a service that handles the actual HTTP delivery of webhook payloads with retry logic, signature generation, and delivery tracking.

## Implementation Details

### File Location
`omoi_os/services/webhook_delivery.py`

### Class Structure
```python
class WebhookDeliveryService:
    def __init__(self, db: DatabaseService):
        self.db = db
        self.http_client = httpx.AsyncClient(timeout=30.0)

    async def deliver(
        self,
        subscription: WebhookSubscription,
        event_type: str,
        payload: dict,
    ) -> WebhookDelivery:
        """Deliver webhook with retry logic."""
        ...

    def _generate_signature(self, payload: bytes, secret: str) -> str:
        """Generate HMAC-SHA256 signature."""
        ...

    async def _attempt_delivery(
        self,
        url: str,
        payload: dict,
        headers: dict,
    ) -> tuple[bool, int, str]:
        """Single delivery attempt. Returns (success, status_code, response)."""
        ...

Retry Configuration

• Max retries: 3
• Backoff: exponential with jitter
• Delays: 1s, 2s, 4s (base)
• Timeout per request: 30 seconds

Headers to Include

X-OmoiOS-Event: {event_type}
X-OmoiOS-Delivery: {delivery_id}
X-OmoiOS-Signature: sha256={signature}
Content-Type: application/json

Verification

# Run unit tests
uv run pytest tests/unit/services/test_webhook_delivery.py -v

# Test signature generation
uv run python -c "
from omoi_os.services.webhook_delivery import WebhookDeliveryService
svc = WebhookDeliveryService(None)
sig = svc._generate_signature(b'{\"test\": true}', 'secret')
print(f'Signature: {sig}')
"

Done When

• WebhookDeliveryService class implemented
• Signature generation matches standard HMAC-SHA256
• Retry logic uses exponential backoff with jitter
• Each delivery attempt is recorded in WebhookDelivery table
• Unit tests pass with 100% coverage of retry scenarios

### Running the Complete Flow

After creating your spec files, run these commands:

```bash
# Navigate to the scripts directory
cd /root/.claude/skills/spec-driven-dev/scripts

# Step 1: Validate all specs (checks frontmatter, references, structure)
python spec_cli.py validate

# Expected output:
# ✓ Validation passed!
# - 2 tickets validated
# - 6 tasks validated
# - All dependencies resolved
# - All references valid

# Step 2: Preview what will be synced (dry run)
python spec_cli.py sync diff

# Expected output:
# Connecting to https://api.omoios.dev...
# Connected!
#
# Sync Results:
# ------------------------------------------------------------
# [CREATE] ticket TKT-001
#          Would create: Webhook Infrastructure Setup
# [CREATE] ticket TKT-002
#          Would create: Platform-Specific Formatters
# [CREATE] task TSK-001
#          Would create: Add webhook models to database
# ... (more tasks)
# ------------------------------------------------------------
# Summary: 8 would be created, 0 updated, 0 skipped

# Step 3: Actually sync to the API
python spec_cli.py sync push

# Expected output:
# Connecting to https://api.omoios.dev...
# Connected!
# Syncing to API...
#
# Sync Results:
# ------------------------------------------------------------
# [CREATE] ticket TKT-001
#          Created with API ID: 550e8400-e29b-41d4-a716-446655440000
# [CREATE] task TSK-001
#          Created with API ID: 550e8400-e29b-41d4-a716-446655440001
# ... (more items)
# ------------------------------------------------------------
# Summary: 8 created, 0 updated, 0 skipped, 0 failed

# Step 4: Verify what's in the project
python spec_cli.py project

# Expected output:
# ══════════════════════════════════════════════════════════════
# PROJECT: My Project
# ══════════════════════════════════════════════════════════════
#
# TICKETS (2 total)
# ──────────────────────────────────────────────────────────────
# • TKT-001: Webhook Infrastructure Setup [backlog]
#   └── 6 tasks
# • TKT-002: Platform-Specific Formatters [backlog]
#   └── 2 tasks

Common Issues & Solutions

Issue: "Validation failed - missing frontmatter"

# Check which files are missing frontmatter
python spec_cli.py validate

# Solution: Add YAML frontmatter block at the top of each file
# Every file MUST start with --- and end the frontmatter with ---

Issue: "Dependency not found: TSK-099"

# Your task references a dependency that doesn't exist
# Solution: Either create TSK-099.md or remove it from depends_on

Issue: "Connection refused" or timeout

# Check if API URL is correct
echo $OMOIOS_API_URL

# In sandbox, this should be auto-injected
# If empty, the CLI falls back to https://api.omoios.dev

Directory Structure Summary

.omoi_os/
├── docs/                    # PRDs and high-level documents
│   └── prd-{feature}.md
├── requirements/            # EARS-format requirements
│   └── {feature}.md
├── designs/                 # Technical designs
│   └── {feature}.md
├── tickets/                 # Parent work items
│   ├── TKT-001.md
│   └── TKT-002.md
└── tasks/                   # Atomic work units
    ├── TSK-001.md
    ├── TSK-002.md
    └── ...

CLI Reference

View Local Specs

python spec_cli.py show all           # Everything
python spec_cli.py show requirements  # Just requirements
python spec_cli.py show designs       # Just designs
python spec_cli.py show tickets       # Just tickets
python spec_cli.py show tasks         # Just tasks
python spec_cli.py show graph         # Task dependency graph
python spec_cli.py show ticket-graph  # Ticket dependency graph
python spec_cli.py show traceability  # Full traceability matrix
python spec_cli.py show ready         # Tasks ready to work on

Validate

python spec_cli.py validate           # Check for issues

Sync to API

python spec_cli.py projects                              # List projects
python spec_cli.py sync diff --project-id <uuid>         # Preview changes
python spec_cli.py sync push --project-id <uuid>         # Push to API
python spec_cli.py project <uuid>                        # View project in API
python spec_cli.py api-trace <uuid>                      # Full API traceability

🔒 Sandbox Requirements

Environment Variables

The orchestrator injects these environment variables into each sandbox session:

Required Imports

Each sandbox MUST have access to:

# Core imports for spec generation
from omoi_os.schemas.spec_generation import (
    SpecPhase,
    ExplorationContext,
    RequirementsOutput,
    DesignOutput,
    TasksOutput,
)
from omoi_os.services.spec_state_machine import SpecStateMachine
from omoi_os.evaluators import (
    ExplorationEvaluator,
    RequirementEvaluator,
    DesignEvaluator,
    TaskEvaluator,
)

File System Access

Sandboxes need read/write access to:

./                           # Project root
./.omoi_os/                  # Spec artifacts directory
./.omoi_os/phase_data/       # JSON outputs from each phase
./.omoi_os/docs/             # PRD documents
./.omoi_os/requirements/     # Requirement documents
./.omoi_os/designs/          # Design documents
./.omoi_os/tickets/          # Ticket markdown files
./.omoi_os/tasks/            # Task markdown files

Session Transcript Persistence

For cross-sandbox resumption, transcripts are saved:

{
    "session_id": "session-123-abc",
    "spec_id": "550e8400-...",
    "phase": "REQUIREMENTS",
    "turn_count": 15,
    "messages": [
        {"role": "user", "content": "..."},
        {"role": "assistant", "content": "..."}
    ],
    "phase_data": {
        "explore": {...},
        "requirements": {...}
    },
    "created_at": "2025-01-10T...",
    "updated_at": "2025-01-10T..."
}

Phase Timeout Handling

If a phase times out:

1. Save current progress to .omoi_os/phase_data/{phase}.partial.json
2. Set resumption flag in session transcript
3. Return control to orchestrator for retry or manual intervention
4. On resume: Load partial data and continue from last checkpoint

Best Practices

Discovery

• Never skip questions - The 5 minutes spent asking saves hours of rework
• Get explicit scope boundaries - "Out of scope" is as important as "in scope"
• Identify risks early - What could go wrong?

PRDs

• Write for a human reader who doesn't have context
• Focus on "why" and "what", not "how"
• Keep it concise but complete

Requirements

• Use EARS format consistently
• Make every requirement testable
• Include acceptance criteria

Designs

• Architecture diagrams first
• Include data models with actual SQL/Pydantic
• Specify APIs completely

Tickets

• One component per ticket
• Clear acceptance criteria
• Explicit dependencies

Tasks

• 1-4 hours of work max
• All context included
• Specific file deliverables

Recovering from Lost Context

If you lose context mid-conversation:

# 1. Check what exists locally
ls .omoi_os/
cat .omoi_os/docs/prd-*.md
cat .omoi_os/requirements/*.md
cat .omoi_os/designs/*.md

# 2. View the full state
python spec_cli.py show all

# 3. Check what's in the API
python spec_cli.py project

Then resume from where you left off based on what's complete and what's missing.