A multi-agent orchestration plugin for OpenCode that enables autonomous work, background tasks, and precise project planning.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install oh-my-opencode
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install oh-my-opencode using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Oh My OpenCode?
Oh My OpenCode is a sophisticated multi-agent orchestration plugin designed to elevate the standard coding environment into a high-performance agent harness. By integrating these Openclaw Skills, developers can access a suite of specialized agents—such as Sisyphus for orchestration, Hephaestus for deep work, and Oracle for high-IQ reasoning—to handle complex development workflows. The skill transforms how OpenCode operates, providing a framework for category-based model routing and non-blocking background task execution.
This plugin is particularly valuable for teams and individual developers who need to manage large codebases or complex features that require planning and multi-step execution. Through its support for specialized workflows and model priority chains, it ensures that tasks are routed to the most capable LLM for the specific domain, whether it be visual engineering or deep logical reasoning. By leveraging these Openclaw Skills, users can achieve higher autonomy and efficiency in their daily development routines.
Oh My OpenCode Use Cases
- Implementing complex features autonomously using the Ultrawork mode to explore, research, and code without manual intervention.
- Managing large-scale project migrations or refactors using the Prometheus planning and Atlas orchestration workflow.
- Running documentation searches and codebase pattern exploration in the background while continuing active development.
- Optimizing model costs and performance by routing specific tasks (like frontend styling or documentation) to domain-optimized models.
- Visualizing and managing multi-agent execution threads within a terminal using the integrated tmux support.
How Oh My OpenCode Works
- The user triggers the plugin by including keywords like ulw or using slash commands within the OpenCode interface.
- Sisyphus, acting as the primary orchestrator, analyzes the intent and determines if a specialized sub-agent or category routing is required.
- In Prometheus mode, the agent conducts an interview to gather requirements and generates a detailed execution plan stored in the .sisyphus directory.
- Tasks are delegated to specialized agents (e.g., Librarian for docs, Explore for code search) which can run in parallel background threads to avoid blocking the main workflow.
- The system uses a predefined provider priority chain to resolve the best available model for each specific category of work.
- Atlas orchestrates the final execution of planned tasks, verifying completion and maintaining state across sessions via unique session IDs.
Oh My OpenCode Setup
To install and configure this plugin, ensure you have OpenCode 1.0.150+ installed and at least one LLM provider authenticated. Follow these steps:
# Install the plugin using bunx
bunx oh-my-opencode install
# Verify the installation
opencode --version
# Run the doctor command to check for configuration issues
bunx oh-my-opencode doctor
Oh My OpenCode Data Schema & Taxonomy
The skill organizes its operations and persistence through a specific directory structure and configuration schema to ensure continuity across Openclaw Skills workflows.
| Data Component | Format | Purpose |
|---|---|---|
.sisyphus/plans/*.md |
Markdown | Stores detailed work plans generated by the Prometheus agent. |
oh-my-opencode.json |
JSON | Main configuration for agents, categories, provider chains, and hooks. |
session_id |
String | Unique identifier used to preserve context and save tokens during follow-up tasks. |
AGENTS.md |
Markdown | Hierarchical knowledge base for deep project understanding. |
Oh My OpenCode Advanced Features
- Parallel Agent Execution: Trigger multiple background agents for simultaneous research and exploration using the run_in_background parameter.
- Tmux Integration: Run multi-agent tasks in separate panes for visual monitoring of parallel work streams.
- Custom Hooks: Utilize built-in hooks like todo-continuation-enforcer to ensure agents complete work even if they encounter mid-task interruptions.
- Category-Based Routing: Define custom priority chains for different task types, such as ultrabrain for complex logic or visual-engineering for UI work.
- MCP Server Support: Native integration with Model Context Protocol servers like Exa for web search and Context7 for documentation lookup.
SKILL.md
Loading
Related Openclaw Skills
Snipgrapher
A command-line utility for generating professional-grade code snippet images in SVG, PNG, and WebP formats.
cEDH Advisor Skill
A high-performance tactical assistant for Competitive Commander (cEDH) that provides instant mana calculations, tutor target optimization, and banlist-compliant play recommendations.
Auto-Improve Skill
A self-optimizing feedback loop that turns every AI interaction into a learning opportunity by tracking mistakes and saving successful patterns.
Anticipation Skill
A proactive intelligence layer that predicts user needs and automates solutions based on behavioral patterns and session history.
cybersec-helper
A professional security assistant designed to guide developers through ethical application reviews, bug bounty recon, and secure coding practices.
Patrick: Executive Expertise & Decision Infrastructure
Patrick is a licensed expertise management system that provides structured, context-aware decision infrastructure for AI agents and executives.
Featured*
