# AIOrchestra — Complete Platform Reference

> AI-powered software development platform where 25 specialized agents collaborate to plan, build, review, test, and deploy production-ready code following a full SDLC process.

Website: https://aiorchestra.dev
Contact: info@aiorchestra.com

---

## Overview

AIOrchestra is a browser-based IDE and multi-agent AI platform. Users describe what they want to build in plain English, and a team of 25 specialized AI agents collaborates through structured sprints to deliver production-ready, reviewed, and tested code.

Unlike single-AI code generators, AIOrchestra runs a full engineering process: product definition, architecture design, development, code review, testing, security audit, and deployment — with dedicated agents for each role that review each other's work.

## How It Works

1. **Describe**: Tell AIOrchestra what you want to build in plain English
2. **Plan**: AI agents define user stories and design the technical architecture
3. **Build**: Developer agents write clean, production-grade code
4. **Verify**: Code reviewers, QA engineers, and security testers validate the work
5. **Deploy**: Publish to a live URL with one click

## AI Agent Team (25 Agents)

### Product & Management
- **Product Owner**: Breaks ideas into user stories with acceptance criteria
- **Product Manager**: Defines product vision, prioritizes features, manages roadmap
- **Scrum Master**: Plans sprints, assigns agents, adapts execution in real-time

### Architecture & Design
- **Solution Architect**: Plans technical architecture, data models, API contracts
- **UI/UX Designer**: Creates interface designs, wireframes, user flows
- **UI/UX Reviewer**: Reviews accessibility, responsiveness, visual consistency

### Development
- **Principal React Developer**: Senior-level React/TypeScript frontend implementation
- **Principal Angular Developer**: Senior-level Angular implementation
- **Principal Vue Developer**: Senior-level Vue.js implementation
- **Backend Node.js Engineer**: Node.js/TypeScript backend services
- **Backend Python Engineer**: Python backend with FastAPI/Django
- **Backend .NET Engineer**: C#/.NET backend services
- **Build & Deploy Engineer**: CI/CD pipelines, build configuration, deployment

### Quality & Security
- **Frontend React Reviewer**: Reviews React code for quality and best practices
- **Frontend Angular Reviewer**: Reviews Angular code quality
- **Frontend Vue Reviewer**: Reviews Vue.js code quality
- **Backend Node.js Reviewer**: Reviews Node.js backend code
- **Backend Python Reviewer**: Reviews Python backend code
- **Backend .NET Reviewer**: Reviews .NET backend code
- **QA Engineer**: Writes and runs tests for APIs, business logic, edge cases
- **Security Tester**: Audits for auth bypasses, injection, input validation
- **API Quality Engineer**: Validates API endpoints, authorization, data integrity

### Marketing & Content
- **Marketing Manager**: Brand development, product positioning, content strategy
- **Content Marketing Strategist**: Website copy, social media, email campaigns
- **Visual Asset Creator**: SVG graphics, icons, illustrations, design assets

## Supported LLM Providers (9 Providers)

| Provider | Models |
|----------|--------|
| Anthropic | Claude Opus, Claude Sonnet, Claude Haiku |
| OpenAI | GPT-4.1, GPT-4o, GPT-4o-mini |
| Google | Gemini 2.5 Pro, Gemini 2.5 Flash |
| DeepSeek | DeepSeek V3, DeepSeek R1 |
| Moonshot (Kimi) | Kimi K2.5 |
| Alibaba (Qwen) | Qwen3 Coder Plus, Qwen3 Coder |
| Mistral | Mistral Large, Mistral Medium |

Smart routing automatically selects the best model for each task based on complexity, cost, and available quotas.

## Key Platform Features

### Sprint Orchestration
An AI Scrum Master orchestrates multi-agent sprints. After each agent step, the Scrum Master evaluates progress and decides the next action — assigning the next agent, requesting rework, or completing the sprint. This adaptive loop ensures quality through iterative review.

### Live Preview
Apps are rendered in real-time as agents write code. The platform auto-detects the tech stack (React, Next.js, Vue, Angular, etc.) and spins up a dev server. Preview URLs are accessible at `/preview/{sessionId}/`.

### One-Click Publish
Deploy any project to a live URL at `{subdomain}.aiorchestra.dev` with automatic TLS via Cloudflare. Supports both static SPAs and server-side rendered apps (Next.js). Published apps get their own PostgreSQL database when needed.

### Template Catalog
Pre-built starter templates:
- React SaaS Starter (Auth, Dashboard, Billing, User Management)
- Next.js E-commerce (Product Catalog, Cart, Checkout)
- Node.js REST API (CRUD, JWT Auth, Validation)
- Python FastAPI (Async API, SQLAlchemy, Pydantic)
- Vue.js Dashboard (Admin Panel, Charts, Analytics)
- React Native Mobile (Cross-platform iOS/Android)

### Full Git Integration
Built-in Git support with branch management, visual diff viewer, merge conflict resolution, and AI-generated commit messages and pull requests.

### Code Review & Security
Every sprint automatically includes:
- Code review for performance, patterns, and correctness
- Security audit for auth flaws, injection, and input validation
- Test validation to verify functionality before merging

## Pricing

### Free — $0/month
- 1 project
- 3 sprints per month
- Free AI models (Qwen, DeepSeek, Kimi)
- 10 welcome Chords + 3 daily Chords for 30 days
- Public previews

### Hobby — $9/month
- 5 projects
- Unlimited sprints
- Sonnet + GPT-4o-mini + Gemini models
- 100 Chords per month

### Pro — $29/month
- Unlimited projects
- All 9 LLM providers including Claude Opus
- 500 Chords per month
- Private previews
- Priority support

### Team — $99/month per seat
- Everything in Pro
- 5 team members included
- 1,500 Chords per seat per month
- Shared workspaces
- Custom agent templates

## Chords (Credit System)
Chords are the platform's credit unit. Each AI interaction consumes Chords based on the model used and tokens processed. Free tier users receive daily Chords; paid plans include monthly allocations.

## Technology Stack

- **Frontend**: React 19, TypeScript, Vite 6, Tailwind CSS, CodeMirror 6
- **Backend**: Fastify 5, tRPC v11, WebSocket streaming
- **Core SDK**: Provider adapters, context engine, smart routing, tool registry
- **Database**: SQLite (platform), PostgreSQL (published apps)
- **Deployment**: Caddy reverse proxy, Cloudflare TLS, custom subdomains

## Comparisons

AIOrchestra vs other AI coding tools:

| Feature | Single AI Chat | AIOrchestra |
|---------|---------------|-------------|
| Architecture | Jumps to code without a plan | Designs architecture before coding |
| Code Quality | Ships first draft, no review | Every change reviewed for correctness |
| Testing | Skips tests or writes stubs | Writes real tests, validates they pass |
| Security | No security analysis | Audits for auth flaws, injection, input abuse |
| Process | One long unstructured chat | Organized sprints with planning, execution, review |
| Iteration | Manual fix requests | Agents review each other and iterate automatically |

Comparison pages:
- vs Cursor: https://aiorchestra.dev/compare/cursor
- vs Bolt: https://aiorchestra.dev/compare/bolt
- vs Lovable: https://aiorchestra.dev/compare/lovable
- vs v0: https://aiorchestra.dev/compare/v0
- vs Replit: https://aiorchestra.dev/compare/replit

## URLs

- Landing Page: https://aiorchestra.dev/landing
- Sign Up: https://aiorchestra.dev/register
- Sign In: https://aiorchestra.dev/login
- Templates: https://aiorchestra.dev/templates
- Contact: https://aiorchestra.dev/contact
- White Paper: https://aiorchestra.dev/whitepaper
- Investors: https://aiorchestra.dev/investors
- Privacy Policy: https://aiorchestra.dev/privacy
- Terms of Service: https://aiorchestra.dev/terms
- Sitemap: https://aiorchestra.dev/sitemap.xml