Configuration Reference

Oh My OpenAgent is highly opinionated but adjustable to taste. Most users don't need to configure anything — run bunx oh-my-opencode install and go.

Quick Start

{
  "$schema": "https://raw.githubusercontent.com/code-yeongyu/oh-my-opencode/master/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" }
  }
}

Config File Locations

.opencode/oh-my-opencode.json (Project level)
~/.config/opencode/oh-my-opencode.json (User level)

JSONC is supported, allowing comments and trailing commas.

Agents

Configure specific behaviors for the built-in agents: Sisyphus, Hephaestus, Oracle, Librarian, Explore, Multimodal Looker, Prometheus, Metis, Momus, Atlas, and Frontend Engineer.

Override Options

Option	Type	Default	Description
model	string	-	Model identifier (e.g., openai/gpt-4o)
variant	string	-	Model variant (max, high, medium, low)
category	string	-	Inherit configuration from a category
temperature	number	-	Sampling temperature (0-2)
top_p	number	-	Top-p sampling (0-1)
prompt	string	-	Override the system prompt completely
prompt_append	string	-	Append text to the system prompt
tools	Record	-	Enable or disable specific tools
disable	boolean	false	Disable this agent
maxTokens	number	-	Maximum tokens for response
thinking	object	-	Extended thinking configuration
reasoningEffort	string	-	Reasoning effort: low, medium, high, xhigh

Permissions

Permission	Values	Description
edit	ask / allow / deny	File editing capabilities
bash	ask / allow / deny	Bash command execution
webfetch	ask / allow / deny	Web request capabilities
doom_loop	ask / allow / deny	Infinite loop override
external_directory	ask / allow / deny	Access files outside project

Skills

Built-in skills include playwright, agent-browser, and git-master. You can also define custom skills.

"skills": {
  "my-custom-skill": {
    "description": "A custom skill for specific tasks",
    "instructions": "Always use this skill when..."
  }
}

Background Tasks

Option	Type	Default	Description
defaultConcurrency	number	-	Default max concurrent tasks
staleTimeoutMs	number	-	Timeout for stale tasks in ms
providerConcurrency	number	-	Concurrency limit per provider
modelConcurrency	number	-	Concurrency limit per model

Priority:modelConcurrency >providerConcurrency >defaultConcurrency

Hooks

Hooks allow you to extend functionality at various lifecycle points.

agent-usage-reminder

anthropic-context-window-limit-recovery

anthropic-effort

atlas

auto-slash-command

auto-update-checker

background-notification

category-skill-reminder

claude-code-hooks

comment-checker

compaction-context-injector

compaction-todo-preserver

delegate-task-retry

directory-agents-injector

directory-readme-injector

edit-error-recovery

interactive-bash-session

keyword-detector

non-interactive-env

prometheus-md-only

question-label-truncator

ralph-loop

rules-injector

session-recovery

sisyphus-junior-notepad

start-work

stop-continuation-guard

subagent-question-blocker

task-reminder

task-resume-info

tasks-todowrite-disabler

think-mode

thinking-block-validator

unstable-agent-babysitter

write-existing-file-guard

MCPs

websearch

context7

Documentation retrieval and context management.

grep_app

GitHub code search integration.

Browser Automation

Tool	Description	Use Case
playwright	Full browser automation (default)	Testing, complex interactions
agent-browser	Lightweight browser agent	Quick lookups, simple scraping

Tmux Integration

Option	Type	Default	Description
enabled	boolean	-	Enable Tmux integration
layout	string	-	Tmux window layout
main_pane_size	string	-	Size of the main pane

Git Master

Option	Type	Default	Description
commit_footer	string	-	Text to append to commit messages
include_co_authored_by	boolean	-	Add Co-authored-by trailer

Comment Checker

Validates comments in your code. Use the {{comments}} placeholder in your custom prompt.

"comment-checker": {
  "custom_prompt": "Review these comments: {{comments}}"
}

Experimental Features

Option	Type	Default	Description
aggressive_truncation	boolean	-	Aggressively truncate outputs
auto_resume	boolean	-	Automatically resume interrupted tasks
preemptive_compaction	boolean	-	Compact context before limits
truncate_all_tool_outputs	boolean	-	Truncate all tool outputs

LSP Configuration

Option	Type	Default	Description
command	string	-	LSP server command
extensions	array	-	File extensions to match
priority	number	-	Server priority
env	object	-	Environment variables
initialization	object	-	Initialization options
disabled	boolean	-	Disable this LSP

Environment Variables

OPENCODE_CONFIG_DIR

Override the default configuration directory path.

Category	Default Model	Description
visual-engineering	gemini-3.1-pro (high)	Frontend, UI/UX, design tasks
ultrabrain	gpt-5.3-codex (xhigh)	Deep logical reasoning
deep	gpt-5.3-codex (medium)	Autonomous problem-solving, thorough research
artistry	gemini-3.1-pro (high)	Creative tasks
quick	claude-haiku-4-5	Trivial, fast tasks
unspecified-low	claude-sonnet-4-6	Low effort general tasks
unspecified-high	gpt-5.4 (high)	High effort general tasks
writing	gemini-3-flash	Documentation and prose