Reference

Packages, config files, CLI commands, categories, hooks, and environment variables.

Package Names

The upstream rename is transitional. Prefer the upstream-published command when installing, and treat OpenAgent names as the public/product identity.

Name	Purpose
`oh-my-opencode`	Current upstream npm package and primary installer command.
`oh-my-openagent`	Compatibility binary/product name used during the rename transition.
`oh-my-openagent` plugin entry	Preferred OpenCode plugin entry when the installer can register it.
`oh-my-opencode` plugin entry	Legacy entry still recognized during transition.

CLI

Command	Purpose
`bunx oh-my-opencode install`	Interactive installer.
`bunx oh-my-opencode install --no-tui ...`	Non-interactive setup for agent-driven installs.
`bunx oh-my-opencode doctor`	System, config, tool, and model-resolution diagnostics.
`bunx oh-my-opencode run <message>`	OpenCode session runner with completion enforcement.
`bunx oh-my-opencode version`	Version information.
`bunx oh-my-opencode mcp oauth ...`	OAuth helper for remote MCP servers.

Config Files

User config is loaded first; project config can override it. JSONC is supported.

Scope	Paths
Project	`.opencode/oh-my-openagent.json[c]`, `.opencode/oh-my-opencode.json[c]`
User	`~/.config/opencode/oh-my-openagent.json[c]`, `~/.config/opencode/oh-my-opencode.json[c]`

Quick Config

{
  "$schema": "https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-opencode.schema.json",
  "agents": {
    "oracle": { "model": "openai/gpt-5.4", "variant": "high" },
    "explore": { "model": "github-copilot/grok-code-fast-1" }
  },
  "categories": {
    "quick": { "model": "opencode/gpt-5-nano" },
    "visual-engineering": { "model": "google/gemini-3.1-pro" }
  }
}

If your installed package exposes a renamed schema, prefer the path reported by doctor.

Agent Options

Option	Meaning
`model`	Explicit `provider/model` override.
`variant`	Variant such as `max`, `high`, `medium`, `low`, or `xhigh`.
`fallback_models`	Ordered fallback chain used when a model/provider fails.
`prompt`, `prompt_append`	Replace or extend an agent prompt; `file://` URIs are supported.
`tools`	Restrict the tools available to an agent.
`disable`	Disable a specific agent.
`permission`	Control permissions such as edit, bash, and webfetch.

Category	Purpose
`visual-engineering`	Frontend, UI/UX, design, styling, animation.
`ultrabrain`	Hard logic, architecture, deep reasoning.
`deep`	Autonomous research plus end-to-end implementation.
`artistry`	Creative or unconventional work.
`quick`	Trivial single-file changes, typo fixes, simple modifications.
`unspecified-low`, `unspecified-high`	Generic low/high-effort fallback categories.
`writing`	Documentation and prose.

Install Flags

Flag	Meaning
`--claude=no\|yes\|max20`	Claude provider mode.
`--openai=no\|yes`	OpenAI/ChatGPT provider mode.
`--gemini=no\|yes`	Gemini provider mode.
`--copilot=no\|yes`	GitHub Copilot fallback provider.
`--opencode-go=no\|yes`	OpenCode Go subscription path.
`--opencode-zen=no\|yes`	`opencode/` model catalog path.
`--zai-coding-plan=no\|yes`	Z.ai Coding Plan path.
`--kimi-for-coding=no\|yes`	Kimi for Coding path.
`--vercel-ai-gateway=no\|yes`	Vercel AI Gateway path.
`--skip-auth`	Skip post-install auth guidance.

Environment

Variable	Meaning
`OMO_SEND_ANONYMOUS_TELEMETRY=0`	Disable anonymous active-install telemetry.
`OMO_DISABLE_POSTHOG=1`	Disable PostHog telemetry.
`OPENCODE_CONFIG_DIR`	Override OpenCode config directory.