Basic Usage
This guide covers the essential shortcuts, commands, and menu options for working with omx.
Keyboard Shortcuts
| Shortcut | Action |
|---|---|
| Escape | Open menu (when idle) / Interrupt generation (when loading) |
| Ctrl+C | Press twice to exit |
| Ctrl+A | Move cursor to line start |
| Ctrl+E | Move cursor to line end |
| Ctrl+L | Clear input and attachments |
| Shift+Enter | Insert newline |
| Ctrl+J | Insert newline |
| Up/Down | Move cursor in multi-line input or browse input history |
Slash Commands
Type these commands directly in the chat:
| Command | Description |
|---|---|
| /clear | Start a fresh conversation |
| /status | Show session, model and configuration status |
| /compact | Manually compact context and start fresh |
| /rewind | Rewind to a previous message |
| /model | Switch to a different model |
| /session | Load a previous session |
| /git-commit | Generate a commit message for staged changes and commit |
| /exit | Exit omx |
When you start typing /, a picker appears with matching commands. Use arrow keys to navigate and Enter to select. Custom skills also appear in this picker.
Menu
Press Escape to open the menu. From here you can:
- Switch to a different model
- Load a previous session
- Rewind to a previous message
- Switch workflow preset
- Manage your model list
- Change preferences
Workflow Presets
omx ships with two workflow presets:
- Programming (default) - Coding assistant with base tools, agent tools, and MCP integration. Concise output, minimal overhead. Reads OMX.md and CLAUDE.md automatically.
- General - Personal assistant for documents, spreadsheets, and presentations. Proactive with tools, conversational tone.
You can create custom workflows by adding markdown files to ~/.omx/workflows/ or .omx/workflows/. See Workflows for details.
Preferences
The preferences submenu lets you toggle:
- Cache duration - How long prompt caching lasts (5 minutes or 1 hour)
- Completion notification - Notify when response takes over a minute
- Context editing - Trim old tool calls and thinking blocks to save tokens
- Cross-session memory - Remember key points across sessions
- Diff view - Show inline diffs when the model edits files
- Extended thinking - Enable deeper reasoning for complex tasks
- IDE context - Include context from your IDE
- Response language - Choose the language for model responses (auto, English, Chinese)
- Server compaction - Let the server handle context compaction instead of the client
- Color theme - Pick a color scheme from 17 presets
- Streaming output - Show responses as they generate
Use arrow keys to navigate, Enter to select, and Escape to go back.