Oh My ClaudeCode

https://omc.vibetip.help/docs
Admin
Oh My ClaudeCode is a multi-agent orchestration layer for Claude Code that coordinates 20...
Tokens:51,893
Snippets:586
Trust Score:2.1
Update:6 months ago
Show doc for...
Tokens:
Raw
### OMC Setup Command Usage

Source: https://omc.vibetip.help/docs/skills/utility/setup

Demonstrates the various ways to invoke the unified setup skill. These commands route to different underlying utilities for installation, diagnostics, or configuration.

```bash
/oh-my-claudecode:setup
/oh-my-claudecode:setup doctor
/oh-my-claudecode:setup mcp
/oh-my-claudecode:setup wizard
/oh-my-claudecode:setup --local
/oh-my-claudecode:setup --global
```

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

### Setup Skill - Compatibility and Related Skills

Source: https://omc.vibetip.help/docs/skills/utility/setup

Explains the compatibility of the Setup skill with individual sub-skills and lists related skills for further configuration.

```APIDOC
## Setup Skill - Compatibility and Related Skills

### Description
This section covers the compatibility of the Setup skill with other OMC skills and lists related skills for advanced configuration.

### Compatibility
`omc-setup`, `omc-doctor`, and `mcp-setup` can still be invoked individually. However, new documentation recommends using the unified `/oh-my-claudecode:setup` entry point.

### Related Skills
* **omc-setup**: Full OMC setup wizard.
* **omc-doctor**: Installation issue diagnostics.
* **mcp-setup**: MCP server configuration.
```

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

### Setup Skill - Overview and Usage

Source: https://omc.vibetip.help/docs/skills/utility/setup

Provides an overview of the Setup skill and demonstrates its various usage commands for different configuration tasks.

```APIDOC
## Setup Skill - Overview and Usage

### Description
The Setup skill consolidates OMC's configuration-related skills into a single entry point. It routes to the appropriate sub-skill based on the arguments provided.

### Usage Examples
```
/oh-my-claudecode:setup              # Full setup wizard (→ omc-setup)
/oh-my-claudecode:setup doctor       # Installation diagnostics (→ omc-doctor)
/oh-my-claudecode:setup mcp          # MCP server configuration (→ mcp-setup)
/oh-my-claudecode:setup wizard       # Explicit wizard path (→ omc-setup)
/oh-my-claudecode:setup --local      # Local project configuration (→ omc-setup --local)
/oh-my-claudecode:setup --global     # Global configuration (→ omc-setup --global)
```
```

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

### Setup Skill - Routing Rules

Source: https://omc.vibetip.help/docs/skills/utility/setup

Details the routing logic of the Setup skill, specifying which sub-skill is invoked based on the provided arguments.

```APIDOC
## Setup Skill - Routing Rules

### Description
This section outlines how the Setup skill routes commands to different sub-skills based on the arguments provided.

### Routing Table
| Argument | Routes To |
|---|---|
| none, `wizard`, `local`, `global`, `--force` | `/oh-my-claudecode:omc-setup` |
| `doctor` | `/oh-my-claudecode:omc-doctor` |
| `mcp` | `/oh-my-claudecode:mcp-setup` |
```

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

### Verify MCP Installation

Source: https://omc.vibetip.help/docs/skills/utility/mcp-setup

Lists all configured MCP servers to verify the installation and ensure they are correctly set up.

```bash
claude mcp list
```

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

### Skill Composition Example

Source: https://omc.vibetip.help/docs/concepts

Illustrates how skills can be composed and used together, showing an example task with active skills including 'ultrawork', 'default', and 'git-master'.

```natural_language
Task: "ultrawork: refactor the API and commit it"
Active skills: ultrawork + default + git-master
```

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

### Calculation and Transformation Example

Source: https://omc.vibetip.help/docs/tools/python-repl

Example of performing calculations, such as estimating token costs, using the Python REPL.

```APIDOC
## Calculation and Transformation Example

### Description
This example demonstrates calculating the estimated cost based on input and output tokens.

### Request Example
```json
{
  "tool_code": "python_repl",
  "code": """
# Token cost calculation
input_tokens = 150000
output_tokens = 50000
cost = (input_tokens * 0.003 + output_tokens * 0.015) / 1000
print(f"Estimated cost: ${cost:.4f}")
"""
}
```

### Response Example
```json
{
  "result": "Estimated cost: $0.8250"
}
```
```

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

### Autopilot Skill Example

Source: https://omc.vibetip.help/docs/concepts

Shows an example of using the 'autopilot' skill, which is designed for an autonomous execution pipeline from idea to finished code.

```natural_language
autopilot build me a REST API with authentication
```

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

### Agent Selection Guide

Source: https://omc.vibetip.help/docs/concepts

Provides a guide for selecting the recommended agent based on the task type, along with the corresponding model to be used for optimal performance.

```markdown
Task Type| Recommended Agent| Model  
---|---|---
Quick code lookup| `explore`| haiku  
Feature implementation| `executor`| sonnet  
Complex refactoring| `executor` (model=opus)| opus  
Simple bug fix| `debugger`| sonnet  
Complex debugging| `architect`| opus  
UI component| `designer`| sonnet  
Documentation| `writer`| haiku  
Test strategy| `test-engineer`| sonnet  
Security review| `security-reviewer`| sonnet  
Code review| `code-reviewer`| opus  
Data analysis| `scientist`| sonnet  
```

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

### UltraQA Progress Output Example

Source: https://omc.vibetip.help/docs/skills/workflow/ultraqa

A representation of the console output generated by the UltraQA workflow during its execution cycles.

```text
[ULTRAQA Cycle 1/5] Running tests...
[ULTRAQA Cycle 1/5] FAILED - 3 tests failing
[ULTRAQA Cycle 1/5] Architect diagnosing...
[ULTRAQA Cycle 1/5] Fixing: auth.test.ts - missing mock
[ULTRAQA Cycle 2/5] Running tests...
[ULTRAQA Cycle 2/5] PASSED - All 47 tests pass
[ULTRAQA COMPLETE] Goal met after 2 cycles
```

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

### Debugging Workflow Examples

Source: https://omc.vibetip.help/docs/guides

Common command sequences for handling test failures and production issues using a combination of analysis, tracing, and fixing agents.

```bash
# Test Failure Example
analyze why test_auth.py is failing
/oh-my-claudecode:trace
tdd: fix the failing test

# Production Issue Example
deep-analyze the memory leak in the worker process
/oh-my-claudecode:trace
ralph: fix the memory leak and add regression tests
```

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

### UltraQA Command Usage

Source: https://omc.vibetip.help/docs/skills/workflow/ultraqa

Examples of how to invoke the UltraQA workflow via CLI with various goal flags such as tests, build, lint, and custom patterns.

```bash
/oh-my-claudecode:ultraqa --tests
/oh-my-claudecode:ultraqa --build
/oh-my-claudecode:ultraqa --lint
/oh-my-claudecode:ultraqa --typecheck
/oh-my-claudecode:ultraqa --custom "all endpoints return 200"
/oh-my-claudecode:ultraqa --interactive
```

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

### Invoke Notification Configuration Wizard

Source: https://omc.vibetip.help/docs/skills/utility/configure-notifications

Commands to launch the interactive configuration wizard for specific notification platforms or the general setup tool.

```bash
/oh-my-claudecode:configure-notifications
/oh-my-claudecode:configure-notifications telegram
/oh-my-claudecode:configure-notifications discord
/oh-my-claudecode:configure-notifications slack
```

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

### Ralplan Skill Example

Source: https://omc.vibetip.help/docs/concepts

Provides an example of the 'ralplan' skill, which involves a planning process where Planner, Architect, and Critic iterate until consensus is reached.

```natural_language
ralplan this feature
```

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

### UltraQA Usage

Source: https://omc.vibetip.help/docs/skills/workflow/ultraqa

Examples of how to invoke the UltraQA workflow with different goal types.

```APIDOC
## UltraQA Usage Examples

### Description
These commands demonstrate how to initiate the UltraQA workflow with various predefined goals or custom checks.

### Method
CLI Command

### Endpoint
`/oh-my-claudecode:ultraqa`

### Parameters
#### Query Parameters
- **--tests** (flag) - Optional - Runs the QA workflow to ensure all test suites pass.
- **--build** (flag) - Optional - Runs the QA workflow to ensure the build succeeds (exit code 0).
- **--lint** (flag) - Optional - Runs the QA workflow to ensure there are no lint errors.
- **--typecheck** (flag) - Optional - Runs the QA workflow to ensure there are no TypeScript errors.
- **--custom "pattern"** (string) - Optional - Runs the QA workflow to check for a specific pattern in the output. Replace `"pattern"` with your custom check.
- **--interactive** (flag) - Optional - Initiates an interactive mode for CLI/service testing via qa-tester.

### Request Example
```bash
/oh-my-claudecode:ultraqa --tests
/oh-my-claudecode:ultraqa --custom "all endpoints return 200"
```

### Response
#### Success Response (200)
- **Output** (string) - Logs the progress of each cycle and the final outcome (completion or stoppage).

#### Response Example
```
[ULTRAQA Cycle 1/5] Running tests...
[ULTRAQA Cycle 1/5] FAILED - 3 tests failing
[ULTRAQA Cycle 1/5] Architect diagnosing...
[ULTRAQA Cycle 1/5] Fixing: auth.test.ts - missing mock
[ULTRAQA Cycle 2/5] Running tests...
[ULTRAQA Cycle 2/5] PASSED - All 47 tests pass
[ULTRAQA COMPLETE] Goal met after 2 cycles
```
```

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

### HUD Configuration Settings

Source: https://omc.vibetip.help/docs/skills/utility/hud

Example JSON configuration for the HUD skill, defining the active preset and individual element visibility.

```json
{
  "omcHud": {
    "preset": "focused",
    "elements": {
      "omcLabel": true,
      "contextBar": true,
      "agents": true,
      "todos": true,
      "ralph": true,
      "showTokens": false
    }
  }
}
```

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

### Data Analysis Example

Source: https://omc.vibetip.help/docs/tools/python-repl

Example of using the Python REPL for data analysis by reading and processing JSON state files.

```APIDOC
## Data Analysis Example

### Description
This example shows how to read a JSON state file and extract information like the number of stages and the status.

### Request Example
```json
{
  "tool_code": "python_repl",
  "code": """
import json
with open('.omc/research/session-1/state.json') as f:
    state = json.load(f)
print(f"Stages: {len(state['stages'])}")
print(f"Status: {state['status']}")
"""
}
```

### Response Example
```json
{
  "result": "Stages: 5\nStatus: completed"
}
```
```

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

### File Processing Example

Source: https://omc.vibetip.help/docs/tools/python-repl

Example of using the Python REPL for file processing, such as analyzing file extensions within a project directory.

```APIDOC
## File Processing Example

### Description
This example walks through the 'src' directory to count files by their extensions.

### Request Example
```json
{
  "tool_code": "python_repl",
  "code": """
import os
import json

# Project file statistics
extensions = {}
for root, dirs, files in os.walk('src'):
    for f in files:
        ext = os.path.splitext(f)[1]
        extensions[ext] = extensions.get(ext, 0) + 1

for ext, count in sorted(extensions.items(), key=lambda x: -x[1]):
    print(f"{ext}: {count} files")
"""
}
```

### Response Example
```json
{
  "result": ".py: 15 files\n.md: 10 files\n.json: 5 files"
}
```
```

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

### Skill CLI Usage Examples

Source: https://omc.vibetip.help/docs/skills/utility/skill

Demonstrates the various commands available for managing skills via the command-line interface. These commands cover operations like listing, adding, removing, editing, searching, and syncing skills.

```bash
/oh-my-claudecode:skill list
/oh-my-claudecode:skill add [name]
/oh-my-claudecode:skill remove <name>
/oh-my-claudecode:skill edit <name>
/oh-my-claudecode:skill search <query>
/oh-my-claudecode:skill info <name>
/oh-my-claudecode:skill sync
/oh-my-claudecode:skill setup
/oh-my-claudecode:skill scan
```

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

### Execute Python Code via REPL

Source: https://omc.vibetip.help/docs/tools/python-repl

Demonstrates the basic usage of the python_repl tool to execute arbitrary Python code. This example imports the json library, parses a string, and prints the resulting dictionary.

```python
python_repl(code="import json; data = json.loads('{\"key\": \"value\"}'); print(data)")
```

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

### Hook Registration Structure Example

Source: https://omc.vibetip.help/docs/concepts

Defines the structure for registering hooks in `hooks.json`. This example shows a hook for the `UserPromptSubmit` event that runs a Node.js script.

```json
{
  "UserPromptSubmit": [
    {
      "matcher": "*",
      "hooks": [
        {
          "type": "command",
          "command": "node scripts/keyword-detector.mjs",
          "timeout": 5
        }
      ]
    }
  ]
}
```

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

### Ralph Skill Example

Source: https://omc.vibetip.help/docs/concepts

Provides an example of the 'ralph' skill, which operates in a loop until the task is verified as complete, ensuring continuous work until completion.

```natural_language
ralph: refactor the authentication module
```

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

### LSP Servers API

Source: https://omc.vibetip.help/docs/tools/lsp

Returns a list of available language servers and their installation status.

```APIDOC
## lsp_servers

### Description
Returns the list of available language servers and their installation status.

### Method
N/A (Function Call)

### Endpoint
N/A (Function Call)

### Parameters
#### Path Parameters
N/A

#### Query Parameters
N/A

#### Request Body
N/A

### Request Example
```python
print(lsp_servers())
```

### Response
#### Success Response (200)
- **servers** (array) - A list of language servers.
  - **name** (string) - The name of the language server.
  - **installed** (boolean) - Whether the server is installed.

#### Response Example
```json
{
  "servers": [
    {
      "name": "typescript-language-server",
      "installed": true
    },
    {
      "name": "pylsp",
      "installed": false
    }
  ]
}
```
```

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

### Manage Language Servers with lsp_servers

Source: https://omc.vibetip.help/docs/tools/lsp

Checks the installation status and availability of configured language servers.

```python
lsp_servers()
```

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

### State Persistence Example

Source: https://omc.vibetip.help/docs/tools/python-repl

Demonstrates how the Python REPL preserves state across multiple calls, allowing variables, functions, and imports to persist.

```APIDOC
## State Persistence

Variables, functions, and imports defined once remain available in subsequent calls.

### Request Example (First Call)
```json
{
  "tool_code": "python_repl",
  "code": "import pandas as pd; df = pd.read_csv('data.csv')"
}
```

### Request Example (Second Call)
```json
{
  "tool_code": "python_repl",
  "code": "print(df.describe())"
}
```

### Description
In the second call, the `df` variable defined in the first call is preserved and can be used directly.
```

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

### Ultrawork Skill Example

Source: https://omc.vibetip.help/docs/concepts

Demonstrates the 'ultrawork' skill, used for running multiple agents simultaneously with maximum parallelism to implement complex tasks.

```natural_language
ultrawork implement user authentication with OAuth
```

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

### CCG Skill Example

Source: https://omc.vibetip.help/docs/concepts

Shows an example of the 'ccg' (Claude-Codex-Gemini) skill, which queries Codex and Gemini simultaneously, and then Claude synthesizes the results for tasks like code review.

```natural_language
ccg: review this authentication implementation
```

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

### System Reminder Injection Example

Source: https://omc.vibetip.help/docs/concepts

Demonstrates how hooks inject context into Claude using \`<system-reminder>\` tags. The content of the tag dictates the action or provides information.

```xml
<system-reminder>
hook success: Success
</system-reminder>
```

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

### Common Usage Patterns for Project Memory

Source: https://omc.vibetip.help/docs/tools/project-memory

Examples of common workflows including registering project-wide rules, documenting codebase architecture, and recording technical learnings.

```python
# Registering Project Rules
project_memory_add_directive("This repository uses conventional commits")
project_memory_add_directive("Files in src/generated/ must not be manually edited")

# Recording Codebase Structure
project_memory_add_note("API layer: src/api/ → src/services/ → src/repositories/")
project_memory_add_note("Auth: JWT + passport.js, implemented in src/auth/")

# Recording Learnings
project_memory_add_note("tsconfig paths settings need to be synchronized with jest.config")
```

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

### Execute Ralph with Flags

Source: https://omc.vibetip.help/docs/skills/workflow/ralph

Examples of running Ralph with specific configuration flags to skip PRD generation or change the completion reviewer.

```bash
ralph fix all failing tests in the auth module
ralph --no-prd fix the typo in README.md
ralph --critic=codex implement the user profile page
```

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

### HUD Command Interface

Source: https://omc.vibetip.help/docs/skills/utility/hud

Lists the available command-line interface commands for managing the HUD display modes and installation status.

```bash
/oh-my-claudecode:hud              # Show current status (auto-configure if needed)
/oh-my-claudecode:hud setup        # Install/repair the HUD status line
/oh-my-claudecode:hud minimal      # Minimal display
/oh-my-claudecode:hud focused      # Focused display (default)
/oh-my-claudecode:hud full         # Full display
/oh-my-claudecode:hud status       # Detailed status
```

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

### Team Skill Example

Source: https://omc.vibetip.help/docs/concepts

Illustrates the 'team' skill, which coordinates multiple Claude agents simultaneously through a 5-stage pipeline: plan, prd, exec, verify, and fix.

```bash
/oh-my-claudecode:team 3:executor "implement a full-stack todo app"
```

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

### notepad_write_priority()

Source: https://omc.vibetip.help/docs/tools/notepad

Writes to the Priority Context section, replacing existing content. This content is always loaded at session start and should be kept under 500 characters.

```APIDOC
## notepad_write_priority(content: str)

### Description
Writes to the Priority Context section. **Replaces existing content.** Always loaded at session start. Keep under 500 characters.

Use for architecture decisions, core constraints, and information that must never be forgotten.

### Method
Notepad Function

### Endpoint
N/A

### Parameters
#### Request Body
- **content** (string) - Required - The content to write to the Priority Context.

### Request Example
```json
{
  "content": "This project uses TypeScript strict mode"
}
```

### Response
#### Success Response (200)
- **status** (string) - Indicates success or failure.

#### Response Example
```json
{
  "status": "success"
}
```
```

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

### Mode State Management Example

Source: https://omc.vibetip.help/docs/hooks/core-hooks

This JSON object represents the state of an execution mode, including active status, start time, prompt, session details, project path, iteration counts, and linking information. It is used by execution mode hooks to maintain the context of ongoing operations.

```json
{
  "active": true,
  "started_at": "2025-01-15T10:30:00Z",
  "prompt": "ultrawork implement auth",
  "session_id": "abc123",
  "project_path": "/path/to/project",
  "iteration": 0,
  "max_iterations": 10,
  "linked_ultrawork": false,
  "last_checked_at": "2025-01-15T10:30:00Z"
}
```

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

### Transitioning from Plan to Execution

Source: https://omc.vibetip.help/docs/guides

Methods to move from a generated plan to actual implementation, including manual execution and automated pipelines.

```bash
# Manual Execution
ralplan "API refactoring"
ultrawork implement the plan in .omc/plans/

# Autopilot Execution
autopilot refactor the authentication system

# Skill Pipeline
deep-interview → omc-plan → autopilot
```

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

### Configure Startup Codebase Mapping

Source: https://omc.vibetip.help/docs/reference/configuration

Controls the automatic generation of a codebase map upon session initialization. Parameters include enabling the feature and setting constraints on file count and directory depth.

```json
{
  "startupCodebaseMap": {
    "enabled": true,
    "maxFiles": 200,
    "maxDepth": 4
  }
}
```

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

### Configure Project Aliases

Source: https://omc.vibetip.help/docs/skills/utility/project-session-manager

Configuration file format for defining project repositories and local paths in ~/.psm/projects.json.

```json
{
  "aliases": {
    "omc": {
      "repo": "Yeachan-Heo/oh-my-claudecode",
      "local": "~/Workspace/oh-my-claudecode",
      "default_base": "main"
    }
  }
}
```

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

### Notepad Overview and Storage

Source: https://omc.vibetip.help/docs/tools/notepad

Provides an overview of the Notepad tool and its storage path.

```APIDOC
## Notepad Overview
A persistent note system that survives context window compaction.

## Storage Path
```
.omc/notepad.md
```
```

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

### Add Exa Web Search MCP Server

Source: https://omc.vibetip.help/docs/skills/utility/mcp-setup

Sets up the Exa Web Search MCP server for enhanced web search capabilities. Requires an EXA_API_KEY for authentication.

```bash
claude mcp add -e EXA_API_KEY=<key> exa -- npx -y exa-mcp-server
```

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

### Optimize Verification Strategy

Source: https://omc.vibetip.help/docs/guides

Demonstrates the preferred approach to verification by prioritizing existing test suites over expensive agent-based testing tools.

```bash
# Bad: using qa-tester when tests exist
qa-tester verify the login feature

# Good: run existing tests
npm test -- --grep "login"
```

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

### Navigate Symbols with lsp_goto_definition and lsp_find_references

Source: https://omc.vibetip.help/docs/tools/lsp

Navigates to symbol definitions or finds all occurrences of a symbol within the project.

```python
lsp_goto_definition(file="src/auth.ts", line=42, character=10)
lsp_find_references(file="src/auth.ts", line=42, character=10)
```

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

### Get Notepad Statistics

Source: https://omc.vibetip.help/docs/tools/notepad

Returns statistics about the notepad, such as the number of entries and total size. This function is useful for monitoring notepad usage.

```python
notepad_stats()
```

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

### Basic and Consensus-Based Planning Commands

Source: https://omc.vibetip.help/docs/guides

Commands for generating execution plans, including basic planning and consensus-driven workflows.

```bash
/oh-my-claudecode:omc-plan "refactor the authentication system"
/oh-my-claudecode:omc-plan --consensus "refactor the authentication system"
/oh-my-claudecode:omc-plan --review "analyze problems with the current plan"
ralplan this feature
/oh-my-claudecode:ralplan "design the payment system"
/oh-my-claudecode:ralplan --deliberate "production database migration"
```

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

### AGENTS.md Hierarchical Tagging Structure

Source: https://omc.vibetip.help/docs/skills/workflow/deepinit

Example of how AGENTS.md files maintain a parent-child relationship across the directory tree using comment tags.

```markdown
/AGENTS.md                        (root, no parent)
├── src/AGENTS.md                 <!-- Parent: ../AGENTS.md -->
│   ├── src/components/AGENTS.md  <!-- Parent: ../AGENTS.md -->
│   └── src/utils/AGENTS.md       <!-- Parent: ../AGENTS.md -->
└── docs/AGENTS.md                <!-- Parent: ../AGENTS.md -->
```

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

### Get Detailed Mode Status

Source: https://omc.vibetip.help/docs/tools/state

Returns comprehensive status information for a specific mode, including file paths and current state content.

```python
state_get_status(mode="ralph")
```

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

### Manage Project Sessions with PSM CLI

Source: https://omc.vibetip.help/docs/skills/utility/project-session-manager

Commands to initialize, list, attach to, and terminate development sessions. These commands support various project reference formats including aliases, URLs, and issue numbers.

```bash
/oh-my-claudecode:project-session-manager review <ref>
/oh-my-claudecode:project-session-manager fix <ref>
/oh-my-claudecode:project-session-manager feature <project> <name>
/oh-my-claudecode:project-session-manager list
/oh-my-claudecode:project-session-manager attach <session>
/oh-my-claudecode:project-session-manager kill <session>
/oh-my-claudecode:project-session-manager cleanup
/oh-my-claudecode:project-session-manager status
```

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

### Save Note Using OMC Skill

Source: https://omc.vibetip.help/docs/concepts

Example of using the OMC note-taking skill to save information to the notepad, which persists across context compactions.

```bash
# Save a note using the skill
/oh-my-claudecode:note "Use /api/v2 for the API endpoint"
```

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

### Add Context7 MCP Server

Source: https://omc.vibetip.help/docs/skills/utility/mcp-setup

Configures the Context7 MCP server to retrieve official documentation for libraries and frameworks. This server does not require an API key.

```bash
claude mcp add context7 -- npx -y @upstash/context7-mcp
```

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

### Add Custom MCP Server (stdio and HTTP)

Source: https://omc.vibetip.help/docs/skills/utility/mcp-setup

Allows adding custom MCP servers, either as a stdio server requiring a command and arguments, or as an HTTP server with a specified URL.

```bash
# stdio server
claude mcp add <name> -- <command> [args...]
```

```bash
# HTTP server
claude mcp add --transport http <name> <url>
```

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

### Optimize Model Configuration

Source: https://omc.vibetip.help/docs/guides

Demonstrates how to selectively assign high-cost models like 'opus' only to complex tasks rather than applying them globally.

```json
// Bad: setting all agents to opus
{
  "agents": {
    "explore": { "model": "opus" },
    "writer": { "model": "opus" },
    "executor": { "model": "opus" }
  }
}

// Good: keep defaults, change only where needed
{
  "agents": {
    "executor": { "model": "opus" }  // only for complex projects
  }
}
```

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

### Trigger Deep Interview with Magic Keywords

Source: https://omc.vibetip.help/docs/skills/workflow/deep-interview

These examples show how to activate the Deep Interview feature using natural language keywords, allowing for a more conversational initiation of the clarification process.

```text
deep interview me about this project idea
ouroboros help me clarify what I want
I have a vague idea about a task management tool
```

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

### Invoke Deep Interview Command

Source: https://omc.vibetip.help/docs/skills/workflow/deep-interview

This command initiates the Deep Interview process by taking a vague idea or description as input. It's the primary way to start the clarification workflow.

```bash
/oh-my-claudecode:deep-interview "vague idea or description"
```

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

### Deep Interview 3-Stage Pipeline

Source: https://omc.vibetip.help/docs/skills/workflow/deep-interview

Illustrates the recommended three-stage pipeline for project development, starting with Deep Interview for clarity, followed by Ralplan for consensus planning, and finally Autopilot for execution.

```text
deep-interview → ralplan → autopilot
  ambiguity ≤ 20%   consensus plan   execution
```

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

### Optimize Agent Mode Selection

Source: https://omc.vibetip.help/docs/guides

Shows how to avoid wasting tokens by using direct edit requests for simple tasks instead of full autopilot pipelines.

```bash
# Bad: using autopilot for a single file edit
autopilot fix the typo in README.md

# Good: direct edit request
Fix the typo in README.md
```

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

### Execute Test Suite

Source: https://omc.vibetip.help/docs/skills/utility/release

Command to run the project's test suite as part of the release checklist.

```shell
npm run test:run
```
Tokens:
Raw