YOLO Developer Documentation
Autonomous multi-agent AI development system that orchestrates specialized AI agents to handle software development from requirements to implementation.
What is YOLO Developer?
YOLO Developer is an autonomous software development system that uses multiple specialized AI agents working together through a LangGraph-based orchestration engine. Each agent has a specific role:
| Agent | Role | Responsibilities |
|---|---|---|
| Analyst | Requirements Engineering | Crystallize vague inputs, detect ambiguities, flag contradictions |
| PM | Product Management | Generate user stories, prioritize backlog, identify dependencies |
| Architect | System Design | Create ADRs, validate architecture, assess technical risks |
| Dev | Development | Generate code, write tests, follow patterns |
| TEA | Test Engineering | Validate coverage, categorize risks, audit testability |
| SM | Scrum Master | Orchestrate sprints, mediate conflicts, manage handoffs |
Key Features
Multi-Agent Orchestration
The agents communicate through a shared state machine, passing context and decisions between each other. The SM agent monitors health, detects circular logic, and escalates to humans when needed.
Quality Gates
Every agent output passes through configurable quality gates that enforce standards for testability, architecture compliance, and definition of done.
Memory & Learning
ChromaDB-backed vector storage enables semantic search across decisions, while pattern learning automatically detects your codebase conventions.
Brownfield Support
Scan existing projects to detect language, frameworks, structure, and testing setup, then generate .yolo/project-context.yaml for agent guidance.
GitHub Automation
Manage branches, commits, PRs, issues, and releases directly from YOLO Developer.
Issue Import
Convert GitHub issues into structured user stories for sprint planning.
Full Observability
Complete audit trail with decision logging, token cost tracking, and requirement traceability from seed to implementation.
Interactive Gathering
Run guided Q&A sessions to crystallize requirements before seeding.
Interactive Chat
Start a conversational CLI session by running yolo with no arguments.
Web Dashboard
Use the local web UI to monitor sprint status and agent activity.
Quick Example
# Initialize a project
yolo init --name my-api
# Seed requirements
yolo seed requirements.md
# Run autonomous development
yolo run
# Check progress
yolo status
Output:
Sprint Status: IN_PROGRESS
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Stories: 3/8 completed
Current: [DEV] Implementing user authentication endpoint
Agent Activity:
✓ Analyst: Crystallized 12 requirements (2m 34s)
✓ PM: Generated 8 user stories (1m 12s)
✓ Architect: Created 3 ADRs, validated 12-Factor (3m 45s)
→ Dev: Implementing story US-003 (in progress)
○ TEA: Pending
○ SM: Monitoring
Quality Gates: 4/4 passing
Token Usage: 45,230 tokens ($0.42)
Documentation Sections
Installation Guide
Complete installation instructions for all platforms and environments.
CLI Reference
Detailed documentation for all CLI commands with examples and options.
MCP Integration
Guide to using YOLO Developer with Claude Code and other MCP clients.
Python SDK
Programmatic API reference with code examples.
Configuration
All configuration options, environment variables, and best practices.
Architecture
Deep dive into agents, orchestration, memory, and quality gates.
Brownfield Guide
Scan and integrate existing projects with generated context.
GitHub Automation
Manage GitHub workflow automation from YOLO Developer.
Issue Import
Import GitHub issues and generate user stories.
Interactive Gathering
Guided requirements elicitation sessions.
Web Dashboard
Local web UI for sprint visualization.
System Requirements
| Requirement | Minimum | Recommended |
|---|---|---|
| Python | 3.10 - 3.13 | 3.12 - 3.13 |
| Memory | 4 GB | 8 GB |
| Disk | 500 MB | 2 GB (with memory persistence) |
| OS | macOS, Linux, Windows (WSL2) | macOS, Linux |
Roadmap
Current Status
| Epic | Status | Description |
|---|---|---|
| 1-13 | ✅ Complete | Core infrastructure, all agents, CLI, SDK |
| 14 | ✅ Complete | MCP integration + Codex compatibility |
| 2 | ✅ Complete | Brownfield project support |
| 12 | ✅ Complete | GitHub repository management |
Recently Completed
- #8 ChatGPT Codex Support (OpenAI/Codex provider + hybrid routing)
- MCP integration tools, walkthroughs, and audit access
- #14 Interactive requirements gathering sessions
- #3 Web interface with dashboard UI
- #7 Sprint visualization dashboard
Planned Features
LLM Providers
| Issue | Feature | Description |
|---|---|---|
| #1 | Local LLM Support | Ollama, LM Studio, vLLM integration with hybrid routing |
IDE Integrations
| Issue | Feature | Description |
|---|---|---|
| #9 | Cursor IDE Support | VS Code extension with MCP integration for Cursor |
| #10 | GitHub Copilot Support | @yolo chat participant and Copilot Workspace integration |
User Interfaces
| Issue | Feature | Description |
|---|---|---|
| (complete) | Web Dashboard | Local UI with REST API, WebSocket updates, and sprint visualization |
Core Enhancements
| Issue | Feature | Description |
|---|---|---|
| #6 | Plugin System | Create and integrate custom agents into the workflow |
| #11 | Course Correction | Mid-sprint requirement changes with impact analysis |
| #13 | Issue Import | Convert GitHub issues to user stories for development |
Performance
| Issue | Feature | Description |
|---|---|---|
| #4 | Token Efficiencies | Context optimization and deduplication for reduced costs |
| #5 | Large Codebase Support | Performance optimization for 10,000+ file repositories |
| #15 | Token Limit Scheduler | Automatic rate limit handling with pause/resume for long sprints |
Getting Help
- GitHub Issues: Report bugs or request features
- Discussions: Ask questions and share ideas
Built with the BMad Method for AI-assisted development.