Oh My OpenAgent

https://ohmyopenagent.com/docs
Admin
Oh My OpenAgent is a highly opinionated but adjustable configuration framework for managing AI...
Tokens:33,374
Snippets:434
Trust Score:4.7
Update:4 months ago
Show doc for...
Tokens:
Raw
### Quick Start Configuration Example

Source: https://ohmyopenagent.com/docs

A comprehensive example demonstrating how to configure various agents, categories, and concurrency settings for Oh My OpenAgent. This setup defines different AI models for distinct tasks and manages provider and model-level concurrency limits.

```json
{
  "$schema": "https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-opencode.schema.json",

  "agents": {
    // Main orchestrator: Claude Opus or Kimi K2.6 work best
    "sisyphus": {
      "model": "kimi-for-coding/k2p5",
      "ultrawork": { "model": "anthropic/claude-opus-4-7", "variant": "max" },
    },

    // Research agents: cheap fast models are fine
    "librarian": { "model": "google/gemini-3-flash" },
    "explore": { "model": "github-copilot/grok-code-fast-1" },

    // Architecture consultation: GPT-5.5 or Claude Opus
    "oracle": { "model": "openai/gpt-5.5", "variant": "high" },

    // Prometheus inherits sisyphus model; just add prompt guidance
    "prometheus": {
      "prompt_append": "Leverage deep & quick agents heavily, always in parallel.",
    },
  },

  "categories": {
    // quick - trivial tasks
    "quick": { "model": "opencode/gpt-5-nano" },

    // unspecified-low - moderate tasks
    "unspecified-low": { "model": "anthropic/claude-sonnet-4-6" },

    // unspecified-high - complex work
    "unspecified-high": { "model": "anthropic/claude-opus-4-7", "variant": "max" },

    // writing - docs/prose
    "writing": { "model": "kimi-for-coding/k2p5" },

    // visual-engineering - Gemini dominates visual tasks
    "visual-engineering": {
      "model": "google/gemini-3.1-pro",
      "variant": "high",
    },

    // Custom category for git operations
    "git": {
      "model": "opencode/gpt-5-nano",
      "description": "All git operations",
      "prompt_append": "Focus on atomic commits, clear messages, and safe operations.",
    },
  },

  // Limit expensive providers; let cheap ones run freely
  "background_task": {
    "providerConcurrency": {
      "anthropic": 3,
      "openai": 3,
      "opencode": 10,
      "zai-coding-plan": 10,
    },
    "modelConcurrency": {
      "anthropic/claude-opus-4-7": 2,
      "opencode/gpt-5-nano": 20,
    },
  },

  "experimental": { "aggressive_truncation": true, "task_system": true },
  "tmux": { "enabled": false },
}
```

--------------------------------

### Example: Install with no subscriptions

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user has no subscriptions.

```bash
bunx oh-my-openagent install --no-tui --claude=no --gemini=no --copilot=no

```

--------------------------------

### Example: Install with all native subscriptions

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user has all native subscriptions.

```bash
bunx oh-my-openagent install --no-tui --claude=max20 --openai=yes --gemini=yes --copilot=no

```

--------------------------------

### Example: Install with only OpenCode Go subscription

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user only has an OpenCode Go subscription.

```bash
bunx oh-my-openagent install --no-tui --claude=no --openai=no --gemini=no --copilot=no --opencode-go=yes

```

--------------------------------

### Interactive Installation

Source: https://ohmyopenagent.com/docs

Launches the interactive setup wizard for initial configuration.

```bash
bunx oh-my-openagent install
```

--------------------------------

### Fetch Installation Guide with Curl

Source: https://ohmyopenagent.com/docs

Use `curl` to fetch the installation guide, ensuring all critical flags and details are preserved. This method is preferred over WebFetch for LLM agents.

```bash
curl -fsSL https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/heads/dev/docs/guide/installation.md

```

--------------------------------

### Example: Install with only Claude subscription

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user only has a Claude subscription.

```bash
bunx oh-my-openagent install --no-tui --claude=yes --gemini=no --copilot=no

```

--------------------------------

### Example: Install with only OpenCode Zen subscription

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user only has an OpenCode Zen subscription.

```bash
bunx oh-my-openagent install --no-tui --claude=no --gemini=no --copilot=no --opencode-zen=yes

```

--------------------------------

### Example: Install with only GitHub Copilot subscription

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user only has a GitHub Copilot subscription.

```bash
bunx oh-my-openagent install --no-tui --claude=no --gemini=no --copilot=yes

```

--------------------------------

### Example: Install with Claude and OpenAI subscriptions

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user has both Claude and OpenAI subscriptions.

```bash
bunx oh-my-openagent install --no-tui --claude=yes --openai=yes --gemini=no --copilot=no

```

--------------------------------

### Example: Install with Z.ai for Librarian subscription

Source: https://ohmyopenagent.com/docs

Example of installing Oh My OpenAgent when the user has a Z.ai Coding Plan subscription.

```bash
bunx oh-my-openagent install --no-tui --claude=yes --gemini=no --copilot=no --zai-coding-plan=yes

```

--------------------------------

### Task Prompt Guide - Good Example

Source: https://ohmyopenagent.com/docs

Demonstrates a well-structured task prompt including all 7 essential elements for clarity and effectiveness.

```markdown
> **TASK** : Fix mobile layout breaking issue in `LoginButton.tsx` **CONTEXT** : `src/components/LoginButton.tsx`, using Tailwind CSS **MUST DO** : Change flex-direction at `md:` breakpoint **MUST NOT DO** : Modify existing desktop layout **EXPECTED** : Buttons align vertically on mobile
```

--------------------------------

### Run Interactive Installer

Source: https://ohmyopenagent.com/docs

Execute the interactive installer using Bun for a streamlined installation process. This is the recommended method for setting up the CLI.

```bash
bunx oh-my-openagent install # recommended

```

--------------------------------

### Task Prompt Guide - Bad Example

Source: https://ohmyopenagent.com/docs

Illustrates a vague and unhelpful task prompt that lacks necessary detail for an agent.

```markdown
> "Fix this"
```

--------------------------------

### Install Oh My OpenAgent

Source: https://ohmyopenagent.com/docs

Install and configure oh-my-openagent by following the instructions from the provided URL.

```bash
Install and configure oh-my-openagent by following the instructions here:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/refs/heads/dev/docs/guide/installation.md

```

--------------------------------

### Agent Browser CLI Installation

Source: https://ohmyopenagent.com/docs

Command to install the agent-browser CLI globally using bun.

```bash
bun add -g agent-browser
```

--------------------------------

### Check OpenCode Installation

Source: https://ohmyopenagent.com/docs

Verifies if OpenCode is installed on the system. If not, it prompts the user to install it and provides a reference link.

```bash
if command -v opencode &> /dev/null;
then
    echo "OpenCode $(opencode --version) is installed"
else
    echo "OpenCode is not installed. Please install it first."
    echo "Ref: https://opencode.ai/docs"
fi

```

--------------------------------

### Non-Interactive Install with OpenCode Zen

Source: https://ohmyopenagent.com/docs

Use this command for a non-interactive installation of Oh My OpenAgent, specifically enabling the OpenCode Zen provider. This is useful for automated setups.

```bash
bunx oh-my-openagent install --no-tui --claude=no --openai=no --gemini=no --opencode-zen=yes

```

--------------------------------

### Custom Skill SKILL.md Example

Source: https://ohmyopenagent.com/docs

Example structure for a custom skill's SKILL.md file, including metadata and MCP server configuration.

```yaml
---
name: my-skill
description: My special custom skill
mcp:
  my-mcp:
    command: npx
    args: ["-y", "my-mcp-server"]
---
```

--------------------------------

### Install Oh My OpenAgent CLI

Source: https://ohmyopenagent.com/docs

Installs the Oh My OpenAgent CLI with various subscription flags. Use this command after gathering user's subscription information.

```bash
bunx oh-my-openagent install --no-tui --claude=<yes|no|max20> --gemini=<yes|no> --copilot=<yes|no> [--openai=<yes|no>] [--opencode-go=<yes|no>] [--opencode-zen=<yes|no>] [--zai-coding-plan=<yes|no>] [--kimi-for-coding=<yes|no>] [--vercel-ai-gateway=<yes|no>] [--skip-auth]

```

--------------------------------

### Install Oh My OpenAgent with OpenCode Zen

Source: https://ohmyopenagent.com/docs

Run this command to install Oh My OpenAgent and select 'Yes' when prompted about having access to OpenCode Zen models. This ensures proper configuration for OpenCode Zen providers.

```bash
bunx oh-my-openagent install
# Select your subscriptions (Claude, ChatGPT, Gemini, OpenCode Zen, etc.)
# When prompted: "Do you have access to OpenCode Zen (opencode/ models)?" → Select "Yes"

```

--------------------------------

### Task Creation and Workflow Example

Source: https://ohmyopenagent.com/docs

Demonstrates how to create tasks, list them, and update their status to manage dependencies and parallel execution. Tasks with empty 'blockedBy' arrays run in parallel.

```javascript
TaskCreate({ subject: "Build frontend" }); // T-001
TaskCreate({ subject: "Build backend" }); // T-002
TaskCreate({ subject: "Run integration tests", blockedBy: ["T-001", "T-002"] }); // T-003

TaskList();
// T-001 [pending] Build frontend        blockedBy: []
// T-002 [pending] Build backend         blockedBy: []
// T-003 [pending] Integration tests     blockedBy: [T-001, T-002]

TaskUpdate({ id: "T-001", status: "completed" });
TaskUpdate({ id: "T-002", status: "completed" });
// T-003 now unblocked

```

--------------------------------

### Agent Browser CLI Usage Example

Source: https://ohmyopenagent.com/docs

Example command to use the agent-browser CLI for navigating to a website and extracting content.

```markdown
Use agent-browser to navigate to example.com and extract the main heading
```

--------------------------------

### Session Continuity Example Timeline

Source: https://ohmyopenagent.com/docs

This timeline illustrates how session continuity works with /start-work, showing the transition from initial planning and execution to resuming work in a new session.

```text
Monday 9:00 AM
  └─ @plan "Build user authentication"
  └─ Prometheus interviews and creates plan
  └─ User: /start-work
  └─ Atlas begins execution, creates boulder.json
  └─ Task 1 complete, Task 2 in progress...
  └─ [Session ends - computer crash, user logout, etc.]

Monday 2:00 PM (NEW SESSION)
  └─ User opens new session (agent = Sisyphus by default)
  └─ User: /start-work
  └─ [start-work hook reads boulder.json]
  └─ "Resuming 'Build user authentication' - 3 of 8 tasks complete"
  └─ Atlas continues from Task 3 (no context lost)
```

--------------------------------

### Full Object Entry with All Supported Fields

Source: https://ohmyopenagent.com/docs

This example demonstrates all supported object-style parameters for a fallback model, including variant, reasoning effort, temperature, top_p, maxTokens, and thinking configuration. Use provider-appropriate settings for optimal performance.

```json
{
  "agents": {
    "oracle": {
      "model": "openai/gpt-5.5",
      "fallback_models": [
        {
          "model": "openai/gpt-5.3-codex(low)",
          "variant": "xhigh",
          "reasoningEffort": "high",
          "temperature": 0.3,
          "top_p": 0.9,
          "maxTokens": 8192,
          "thinking": {
            "type": "disabled"
          }
        }
      ]
    }
  }
}

```

--------------------------------

### Git-Master Usage Examples

Source: https://ohmyopenagent.com/docs

Provides example commands for interacting with the git-master skill for committing, rebasing, and querying code history.

```markdown
/git-master commit these changes
/git-master rebase onto main
/git-master who wrote this authentication code?
```

--------------------------------

### Start Work Behavior and Session Continuity

Source: https://ohmyopenagent.com/docs

The /start-work command initiates or resumes work. It checks for boulder.json to determine whether to start fresh or resume an existing plan, ensuring task progress is maintained.

```text
User: /start-work
    ↓
[start-work hook activates]
    ↓
Check: Does .omo/boulder.json exist?
    ↓
    ├─ YES (existing work) → RESUME MODE
    │   - Read the existing boulder state
    │   - Calculate progress (checked vs unchecked boxes)
    │   - Inject continuation prompt with remaining tasks
    │   - Atlas continues where you left off
    │
    └─ NO (fresh start) → INIT MODE
        - Find the most recent plan in .omo/plans/
        - Create new boulder.json tracking this plan
        - Switch session agent to Atlas
        - Begin execution from task 1
```

--------------------------------

### Example Category Configuration

Source: https://ohmyopenagent.com/docs

Demonstrates defining new custom categories, overriding existing ones by changing the model, and configuring advanced settings like thinking models and tool restrictions.

```json
{
  "categories": {
    // 1. Define new custom category
    "korean-writer": {
      "model": "google/gemini-3-flash",
      "temperature": 0.5,
      "prompt_append": "You are a Korean technical writer. Maintain a friendly and clear tone.",
    },

    // 2. Override existing category (change model)
    "visual-engineering": {
      "model": "openai/gpt-5.5",
      "temperature": 0.8,
    },

    // 3. Configure thinking model and restrict tools
    "deep-reasoning": {
      "model": "anthropic/claude-opus-4-7",
      "thinking": {
        "type": "enabled",
        "budgetTokens": 32000,
      },
      "tools": {
        "websearch_web_search_exa": false,
      },
    },
  },
}

```

--------------------------------

### Interactive Bash Terminal Usage Examples

Source: https://ohmyopenagent.com/docs

Provides examples for using the interactive_bash tool to manage tmux sessions. Commands are passed directly as tmux subcommands.

```bash
# Create a new session
interactive_bash(tmux_command="new-session -d -s dev-app")

# Send keystrokes to a session
interactive_bash(tmux_command="send-keys -t dev-app 'vim main.py' Enter")

# Capture pane output
interactive_bash(tmux_command="capture-pane -p -t dev-app")

```

--------------------------------

### Playwright MCP Usage Example

Source: https://ohmyopenagent.com/docs

Example command to use the Playwright skill for navigating to a website and taking a screenshot.

```markdown
/playwright Navigate to example.com and take a screenshot
```

--------------------------------

### Verify OpenCode Version

Source: https://ohmyopenagent.com/docs

Checks if the installed OpenCode version meets the minimum requirement (1.0.150 or higher).

```bash
opencode --version

```

--------------------------------

### Architect Agent Configuration

Source: https://ohmyopenagent.com/docs

Example configuration for an 'Architect' agent that uses pure reasoning for in-depth system architecture analysis.

```yaml
load_skills : []
```

--------------------------------

### Configure Anthropic (Claude) Authentication

Source: https://ohmyopenagent.com/docs

Guides through the interactive terminal process for logging into Anthropic (Claude Pro/Max) via OAuth.

```bash
opencode auth login
# Interactive Terminal: find Provider: Select Anthropic
# Interactive Terminal: find Login method: Select Claude Pro/Max
# Guide user through OAuth flow in browser
# Wait for completion
# Verify success and confirm with user

```

--------------------------------

### Designer Agent Configuration

Source: https://ohmyopenagent.com/docs

Example configuration for a 'Designer' agent that combines visual engineering and UI/UX skills with browser rendering verification.

```yaml
load_skills : ["frontend-ui-ux", "playwright"]
```

--------------------------------

### Enable Tmux Integration for Background Agents

Source: https://ohmyopenagent.com/docs

Configures Oh My OpenAgent to display background agents in separate tmux panes for real-time monitoring. This requires tmux to be installed and enabled in the configuration.

```json
{
  "tmux": {
    "enabled": true,
    "layout": "main-vertical"
  }
}
```

--------------------------------

### Run Your First Task with Oh My OpenAgent

Source: https://ohmyopenagent.com/docs

Execute the 'ultrawork' command to have the agent automatically explore, implement, and verify features in your codebase.

```bash
ultrawork

```

--------------------------------

### Display Help

Source: https://ohmyopenagent.com/docs

Shows the help message for the preferred package name.

```bash
bunx oh-my-openagent
```

--------------------------------

### Load Agent Prompts from Files

Source: https://ohmyopenagent.com/docs

Demonstrates how to load agent system prompts and additional content from external files using file:// URLs in the agent configuration.

```json
{
  "agents": {
    "sisyphus": {
      "prompt": "file:///path/to/custom-prompt.md",
    },
    "oracle": {
      "prompt_append": "file:///path/to/additional-context.md",
    },
  },
  "categories": {
    "deep": {
      "prompt_append": "file:///path/to/deep-category-append.md",
    },
  },
}
```

--------------------------------

### List All Available Models

Source: https://ohmyopenagent.com/docs

Use this command to see all `provider/model` combinations accessible by your current system configuration. Providers are determined by your authentication status and the `models.dev` catalog.

```bash
opencode models

```

--------------------------------

### Built-in Command: /start-work

Source: https://ohmyopenagent.com/docs

Initiates execution based on a Prometheus-generated plan. Accepts an optional plan-name argument.

```bash
/start-work [plan-name]
```

--------------------------------

### Retrieve Background Task Results

Source: https://ohmyopenagent.com/docs

Fetches the output of a previously started background task using its ID.

```python
background_output(task_id="bg_abc123")
```

--------------------------------

### Recommended Stack Configuration (OpenCode Go + OpenAI)

Source: https://ohmyopenagent.com/docs

This configuration uses OpenCode Go for many agents and categories, with OpenAI models for specific high-performance needs. It also includes settings for background task provider concurrency.

```json
{
  "$schema": "https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/dev/assets/oh-my-opencode.schema.json",

  "agents": {
    // Sisyphus: Kimi K2.6 is the top alternative to Claude for orchestration
    "sisyphus": {
      "model": "opencode-go/kimi-k2.6",
      "ultrawork": { "model": "opencode-go/kimi-k2.6" },
    },

    // Hephaestus: needs GPT. ChatGPT Plus gets you here.
    "hephaestus": { "model": "openai/gpt-5.5", "variant": "medium" },

    // Architecture consultation: GPT or Claude Opus
    "oracle": { "model": "openai/gpt-5.5", "variant": "high" },

    // Prometheus inherits Sisyphus behavior
    "prometheus": { "model": "opencode-go/kimi-k2.6" },

    // Atlas also communicative — Kimi works great
    "atlas": { "model": "opencode-go/kimi-k2.6" },

    // Utility agents stay cheap
    "explore": { "model": "opencode-go/qwen3.5-plus" },
    "librarian": { "model": "opencode-go/qwen3.5-plus" },
  },

  "categories": {
    "visual-engineering": { "model": "opencode-go/qwen3.6-plus" },  // Qwen as Gemini alt
    "deep": { "model": "openai/gpt-5.5", "variant": "medium" },
    "ultrabrain": { "model": "openai/gpt-5.5", "variant": "xhigh" },
    "quick": { "model": "openai/gpt-5.4-mini" },
    "unspecified-low": { "model": "opencode-go/kimi-k2.6" },
    "unspecified-high": { "model": "opencode-go/kimi-k2.6" },
    "writing": { "model": "opencode-go/kimi-k2.6" },
  },

  "background_task": {
    "providerConcurrency": {
      "openai": 3,
      "opencode-go": 10,
    },
  },
}

```

--------------------------------

### OpenCode Go Only Configuration (Budget, No GPT)

Source: https://ohmyopenagent.com/docs

Configure agents for a budget-friendly, Go-only environment, omitting GPT models. This configuration prioritizes cost-effectiveness over GPT capabilities.

```json
{
  "agents": {
    "sisyphus": { "model": "opencode-go/kimi-k2.6" },
    "atlas": { "model": "opencode-go/kimi-k2.6" },
    // Omit hephaestus entirely; it needs GPT.
    "oracle": { "model": "opencode-go/glm-5.1" },  // Degraded but functional
    "explore": { "model": "opencode-go/qwen3.5-plus" },
    "librarian": { "model": "opencode-go/qwen3.5-plus" },
  },
  "categories": {
    "visual-engineering": { "model": "opencode-go/qwen3.6-plus" },
    "deep": { "model": "opencode-go/kimi-k2.6" },  // Not ideal — Kimi isn't GPT, but best available
    "unspecified-high": { "model": "opencode-go/kimi-k2.6" },
    "unspecified-low": { "model": "opencode-go/kimi-k2.6" },
    "quick": { "model": "opencode-go/minimax-m2.7" },
    "writing": { "model": "opencode-go/kimi-k2.6" },
  },
}

```

--------------------------------

### Maintainer Agent Configuration

Source: https://ohmyopenagent.com/docs

Example configuration for a 'Maintainer' agent that uses cost-effective models for quick code fixes and commit generation.

```yaml
load_skills : ["git-master"]
```

--------------------------------

### Define Skill-Embedded MCP Server

Source: https://ohmyopenagent.com/docs

Declare an MCP server within a skill's SKILL.md frontmatter. This example shows how to define a Playwright MCP server.

```yaml
---
description: Browser automation skill
mcp:
  playwright:
    command: npx
    args: ["-y", "@anthropic-ai/mcp-playwright"]
---
```

--------------------------------

### Built-in Command: /init-deep

Source: https://ohmyopenagent.com/docs

Initializes a hierarchical AGENTS.md knowledge base within a project. Use [--create-new] to create new files and [--max-depth=N] to limit recursion depth.

```bash
/init-deep [--create-new] [--max-depth=N]
```

--------------------------------

### Display Help (Compatibility Package)

Source: https://ohmyopenagent.com/docs

Shows the help message for the compatibility package name.

```bash
bunx oh-my-opencode
```

--------------------------------

### Built-in Command: /ralph-loop

Source: https://ohmyopenagent.com/docs

Starts a self-referential development loop that runs until task completion. Specify the task as a string argument and optionally set --max-iterations.

```bash
/ralph-loop "Build a REST API with authentication"
/ralph-loop "Refactor the payment module" --max-iterations=50
```

--------------------------------

### Log in to a Provider

Source: https://ohmyopenagent.com/docs

Initiates the login process for a cloud provider. If no provider is specified, an interactive picker will appear. This is necessary if the desired model is not listed as available.

```bash
opencode auth login

```

--------------------------------

### Decision Flow for Task Approach

Source: https://ohmyopenagent.com/docs

This flowchart guides users on selecting the appropriate approach based on task complexity and the need for detailed context explanation or precise execution.

```text
Is it a quick fix or simple task?
  └─ YES → Just prompt normally
  └─ NO  → Is explaining the full context tedious?
              └─ YES → Type "ulw" and let the agent figure it out
              └─ NO  → Do you need precise, verifiable execution?
                         └─ YES → Use @plan for Prometheus planning, then /start-work
                         └─ NO  → Just use "ulw"

```

--------------------------------

### Check OpenCode Configuration File

Source: https://ohmyopenagent.com/docs

Inspects the OpenCode configuration file to ensure 'oh-my-openagent' or 'oh-my-opencode' is listed in the plugins.

```bash
cat ~/.config/opencode/opencode.json

```

--------------------------------

### Conditional Rules with Globs

Source: https://ohmyopenagent.com/docs

Defines conditional rules for code injection using globs to match specific file patterns. This example specifies rules for TypeScript and JavaScript files.

```yaml
---
globs: ["*.ts", "src/**/*.js"]
description: "TypeScript/JavaScript coding rules"
---

- Use PascalCase for interface names
- Use camelCase for function names


```

--------------------------------

### Configure Fallback Models with Per-Model Settings

Source: https://ohmyopenagent.com/docs

Set up fallback models for an agent, supporting both simple string fallbacks and detailed per-model configurations.

```json
{
  "agents": {
    "sisyphus": {
      "model": "anthropic/claude-opus-4-7",
      "fallback_models": [
        // Simple string fallback
        "openai/gpt-5.5",
        // Object with per-model settings
        {
          "model": "google/gemini-3.1-pro",
          "variant": "high",
          "temperature": 0.2
        },
        {
          "model": "anthropic/claude-sonnet-4-6",
          "thinking": { "type": "enabled", "budgetTokens": 64000 }
        }
      ]
    }
  }
}

```

--------------------------------

### Inject AGENTS.md Files

Source: https://ohmyopenagent.com/docs

Demonstrates the directory structure and how AGENTS.md files are automatically injected based on their location relative to the current file and the project root.

```bash
project/
├── AGENTS.md              # Injected first
├── src/
│   ├── AGENTS.md          # Injected second
│   └── components/
│       ├── AGENTS.md      # Injected third
│       └── Button.tsx     # Reading this injects all 3


```

--------------------------------

### Configure Custom Agent Models

Source: https://ohmyopenagent.com/docs

Override the default model an agent uses by specifying it in your plugin configuration file. Existing installations commonly use `oh-my-opencode.json`, while `oh-my-openagent.json` is also recognized.

```json
{
  "agents": {
    "sisyphus": { "model": "kimi-for-coding/k2p5" },
    "prometheus": { "model": "openai/gpt-5.5" }, // Auto-switches to the GPT prompt
  },
}

```

--------------------------------

### File-Based Prompts Configuration

Source: https://ohmyopenagent.com/docs

Illustrates loading agent system prompts and category prompt appends from external files using `file://` URLs. Supports home directory expansion and relative paths.

```json
{
  "agents": {
    "sisyphus": {
      "prompt": "file:///path/to/custom-prompt.md"
    },
    "oracle": {
      "prompt_append": "file:///path/to/additional-context.md"
    }
  },
  "categories": {
    "deep": {
      "prompt_append": "file:///path/to/deep-category-append.md"
    }
  }
}

```

--------------------------------

### Authenticate with OpenCode

Source: https://ohmyopenagent.com/docs

This sequence demonstrates the interactive terminal prompts for authenticating with OpenCode using Google OAuth. It includes optional steps for adding multiple Google accounts for load balancing.

```bash
opencode auth login
# Interactive Terminal: Provider: Select Google
# Interactive Terminal: Login method: Select OAuth with Google (Antigravity)
# Complete sign-in in browser (auto-detected)
# Optional: Add more Google accounts for multi-account load balancing
# Verify success and confirm with user

```

--------------------------------

### Invoke Task with a Specific Category

Source: https://ohmyopenagent.com/docs

Demonstrates how to use the 'task' tool with a specified category to optimize agent behavior for a particular domain. This example uses the 'visual-engineering' category for frontend and UI/UX related tasks.

```javascript
task({
  category: "visual-engineering",
  prompt: "Add a responsive chart component to the dashboard page",
});
```

--------------------------------

### Invoke Prometheus by Switching Agent

Source: https://ohmyopenagent.com/docs

Use this method when starting a new session or when you want a clear separation between planning and execution contexts. It involves pressing Tab to access the agent list and selecting Prometheus.

```text
1. Press Tab at the prompt
2. Select "Prometheus" from the agent list
3. Describe your work: "I want to refactor the auth system"
4. Answer interview questions
5. Prometheus creates plan in .omo/plans/{name}.md
```

--------------------------------

### Experimental Configuration Settings

Source: https://ohmyopenagent.com/docs

Configure experimental features like tool output truncation, context pruning, and task system.

```json
{
  "experimental": {
    "truncate_all_tool_outputs": false,
    "aggressive_truncation": false,
    "auto_resume": false,
    "disable_omo_env": false,
    "task_system": true,
    "dynamic_context_pruning": {
      "enabled": false,
      "notification": "detailed",
      "turn_protection": { "enabled": true, "turns": 3 },
      "protected_tools": [
        "task",
        "todowrite",
        "todoread",
        "lsp_rename",
        "session_read",
        "session_write",
        "session_search"
      ],
      "strategies": {
        "deduplication": { "enabled": true },
        "supersede_writes": { "enabled": true, "aggressive": false },
        "purge_errors": { "enabled": true, "turns": 5 }
      }
    }
  }
}
```

--------------------------------

### Log in to a Specific Provider

Source: https://ohmyopenagent.com/docs

Skips the interactive picker and logs you directly into a specified provider, such as `opencode-go`. This is a shortcut for users who know which provider they want to authenticate with.

```bash
opencode auth login --provider opencode-go

```

--------------------------------

### All Native Stack Configuration (Anthropic + OpenAI + Google)

Source: https://ohmyopenagent.com/docs

This configuration prioritizes the highest quality models from Anthropic, OpenAI, and Google, resulting in higher costs. It maps agents and categories to premium model options.

```json
{
  "agents": {
    "sisyphus": {
      "model": "anthropic/claude-opus-4-7",
      "variant": "max",
    },
    "hephaestus": { "model": "openai/gpt-5.5", "variant": "medium" },
    "oracle": { "model": "openai/gpt-5.5", "variant": "high" },
  },
  "categories": {
    "visual-engineering": { "model": "google/gemini-3.1-pro", "variant": "high" },
    "deep": { "model": "openai/gpt-5.5", "variant": "medium" },
    "unspecified-high": { "model": "anthropic/claude-opus-4-7", "variant": "max" },
  },
}

```

--------------------------------

### Load Prompts from File URIs

Source: https://ohmyopenagent.com/docs

Configure agents to load prompts and prompt appends from local files using 'file://' URIs.

```json
{
  "agents": {
    "sisyphus": {
      "prompt_append": "file:///absolute/path/to/prompt.txt"
    },
    "oracle": {
      "prompt": "file://./relative/to/project/prompt.md"
    },
    "explore": {
      "prompt_append": "file://~/home/dir/prompt.txt"
    }
  },
  "categories": {
    "custom": {
      "model": "anthropic/claude-sonnet-4-6",
      "prompt_append": "file://./category-context.md"
    }
  }
}

```

--------------------------------

### List MCP Servers (Plugin-Injected)

Source: https://ohmyopenagent.com/docs

Demonstrates that plugin-injected MCP servers are not visible when listing static OpenCode configurations. Use 'bunx oh-my-openagent doctor --verbose' for runtime inspection.

```bash
# These are plugin-injected — they will NOT appear here
$ opencode mcp list
No MCP servers configured
```

--------------------------------

### Star the Repository using GitHub CLI

Source: https://ohmyopenagent.com/docs

If the user finds the project helpful, you can ask them to star the repository. This command should only be run if the user explicitly consents. It uses the GitHub CLI to perform a PUT request to the starred endpoint.

```bash
gh api --silent --method PUT /user/starred/code-yeongyu/oh-my-openagent >/dev/null 2>&1 || true

```

--------------------------------

### View CLI Package Version

Source: https://ohmyopenagent.com/docs

Check the version of the Oh My OpenAgent CLI package using the version command.

```bash
bunx oh-my-openagent version
```

--------------------------------

### Invoke Prometheus with @plan Command

Source: https://ohmyopenagent.com/docs

Use this method when you are already in Sisyphus and mid-work for a quick planning interrupt. It's a convenience shortcut that automatically switches to Prometheus.

```text
1. Stay in Sisyphus (default agent)
2. Type: @plan "I want to refactor the auth system"
3. The @plan command automatically switches to Prometheus
4. Answer interview questions
5. Prometheus creates plan in .omo/plans/{name}.md
```
Tokens:
Raw