What is Oh My OpenCode?
Oh My OpenCode is a multi-model agent orchestration harness for OpenCode. It transforms a single AI agent into a coordinated development team that actually ships code. Not locked to Claude. Not locked to OpenAI. Not locked to anyone. Just better results, cheaper models, real orchestration.Quick start
Get running with ultrawork in under 5 minutes
Installation
Complete setup with all providers and configuration
Core concepts
Learn about agents, orchestration, and categories
Configuration
Customize agents, models, and features
The philosophy: breaking free
Anthropic wants you locked in. OpenAI wants you locked in. Everyone wants you locked in. Oh My OpenCode doesn’t play that game. It orchestrates across models, picking the right brain for the right job:- Claude for orchestration
- GPT for deep reasoning
- Gemini for frontend
- Haiku for quick tasks
How it works
Instead of one agent doing everything, Oh My OpenCode uses specialized agents that delegate to each other based on task type. When Sisyphus delegates to a subagent, it doesn’t pick a model name. It picks a category:visual-engineering— Frontend, UI/UX, designdeep— Autonomous research + executionquick— Single-file changes, typosultrabrain— Hard logic, architecture decisions
Key features
Ultrawork mode
Type
ultrawork or ulw. That’s it. The agent handles everything — explores, researches, implements, verifies. Keeps working until done.Parallel execution
Fire 5+ agents simultaneously. Research, implementation, and verification happening at once. Like having 5 engineers instead of 1.
Hash-anchored edits
LINE#ID content hashing validates every edit before applying. Grok Code Fast 1 went from 6.7% to 68.3% success rate just from this change.Intent gate
Classifies your true intent before acting — research, implementation, investigation, fix. Fewer misinterpretations, better results.
LSP + AST tools
Workspace-level rename, go-to-definition, find-references, pre-build diagnostics, AST-aware code rewrites. IDE precision for agents.
Skill-embedded MCPs
Each skill brings its own MCP servers, scoped to the task. Context window stays clean instead of bloating with every tool.
Meet the agents
Sisyphus: The discipline agent
Named after the Greek myth. He rolls the boulder every day. Never stops. Never gives up. Sisyphus is your main orchestrator. Plans, delegates to specialists, and drives tasks to completion with aggressive parallel execution. He doesn’t stop halfway.Recommended models: Claude Opus 4.6, Claude Sonnet 4.6, Kimi K2.5, GLM 5Sisyphus has Claude-optimized prompts. No GPT prompt exists. Claude-family models work best.
Hephaestus: The legitimate craftsman
Named with intentional irony. Anthropic blocked OpenCode from using their API because of this project. So the team built an autonomous GPT-native agent instead. Hephaestus runs on GPT-5.3 Codex. Give him a goal, not a recipe. He explores the codebase, researches patterns, and executes end-to-end without hand-holding.Prometheus: The strategic planner
Prometheus interviews you like a real engineer. Asks clarifying questions. Identifies scope and ambiguities. Builds a detailed plan before a single line of code is touched. Press Tab to enter Prometheus mode, or type@plan "your task" from Sisyphus.
Supporting cast
- Atlas — Executes Prometheus plans, distributes tasks to specialized subagents
- Oracle — Read-only high-IQ consultant for architecture decisions
- Metis — Gap analyzer, catches what Prometheus missed
- Momus — Ruthless reviewer, validates plans against clarity criteria
- Explore — Fast codebase grep using speed-focused models
- Librarian — Documentation and OSS code search
- Multimodal Looker — Vision and screenshot analysis
What’s next?
Get started with ultrawork
Install and run your first ultrawork command in minutes
Complete installation
Full setup guide with all providers and authentication
Learn the architecture
Understand how agents collaborate and delegate
Read the manifesto
Philosophy behind the project and why it exists