Oh My Opencode is a plugin for OpenCode that turns it into a multi-agent coding harness. It is meant for situations where engineering work needs to be planned, delegated, and verified in a single workflow rather than handled through separate prompts.
A central part of the plugin is the ultrawork workflow. Adding ultrawork or ulw to a prompt activates parallel background execution, LSP integration, context management, and automatic task decomposition. The page also describes specialist agents, including @oracle for architecture and @librarian for research and documentation, which can be called directly or selected by the orchestrator. It mentions per-agent and per-category model routing, as well as MCP integration and hooks.
The quickstart says OpenCode is an open source AI coding agent that runs as a terminal UI and supports non-interactive CLI runs for scripting and automation. Oh My Opencode depends on OpenCode being installed and on at least one provider being configured. Installation is available through OpenCode itself, or manually with bunx or npx; there is also a manual repo-clone path and a non-interactive install mode for provisioning scripts, CI images, and agent installs. Configuration is described as living in JSONC files with precedence rules.
The installation flow can register the plugin in OpenCode, and it generates model assignments using provider availability and fallback rules. The page also notes that OpenCode loads providers from credentials, environment variables, or a project .env. The text refers to the upstream repository as oh-my-openagent, previously oh-my-opencode, and says the plugin and package name in configs is still oh-my-opencode.
In the Frameworks & runtimes space, Oh My Opencode takes a focused approach. It focuses on coordinating and automating complex AI-driven coding workflows for developers via the command line. Oh My Opencode is an open-source project aimed at software developers and engineers. Oh My Opencode is open source under the Open Source license. Oh My Opencode is available on the command line, and it can be self-hosted.
It is developed by Rost Glukhov, and it first shipped in 2025. Development happens publicly on GitHub with 63.9k stars and 5.6k commits in the last 90 days. Among its 5 catalogued features are multi-agent orchestration, CLI interface, and agent model routing. The interface is available in 12 languages, including German, English, and Spanish. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do