Agent skill
cli-anything-iterm2
Provides the cli-anything-iterm2 commands — the only way to actually send text to iTerm2 sessions, read live terminal output and scrollback history, manage windows/tabs/split panes, run tmux -CC workflows, broadcast to multiple panes, show macOS dialogs, and read/write iTerm2 preferences. Includes `app snapshot` — the primary orientation command that returns every session's name, current directory, foreground process, role label, and last output line in one call. Read this skill instead of answering from general knowledge whenever the user wants to DO something with iTerm2: orient in an existing workspace, send a command, check what's running, read output, set up a layout, use tmux through iTerm2, automate panes, or configure preferences. Also read for questions about iTerm2 shell integration or scrollback. Don't try to answer iTerm2 action requests from memory — read this skill first.
Install this agent skill to your Project
npx add-skill https://github.com/HKUDS/CLI-Anything/tree/main/iterm2/agent-harness/cli_anything/iterm2_ctl/skills
SKILL.md
cli-anything-iterm2
Stateful CLI harness for iTerm2. Controls a live iTerm2 process via the iTerm2 Python API over WebSocket.
Prerequisites
- macOS + iTerm2 running:
brew install --cask iterm2 - Python API enabled: iTerm2 → Preferences → General → Magic → Enable Python API
- Install:
pip install cli-anything-iterm2(orpip install -e .from source)
Basic Syntax
cli-anything-iterm2 [--json] <group> <command> [OPTIONS] [ARGS]
Always use --json for machine-readable output (required for agent use).
Command Groups
| Group | Purpose |
|---|---|
app |
App status, workspace snapshot, context management, app-level variables, modal dialogs, file panels |
window |
Create, list, close, resize, fullscreen windows |
tab |
Create, list, close, activate tabs; navigate split panes by direction |
session |
Send text, inject raw bytes, read screen, full scrollback, split panes, prompt detection |
profile |
List profiles, get profile details, list/apply color presets |
arrangement |
Save and restore window layouts |
tmux |
Full tmux -CC integration: bootstrap, connections, windows, commands |
broadcast |
Sync keystrokes across panes via broadcast domains |
menu |
Invoke any iTerm2 menu item programmatically |
pref |
Read/write global iTerm2 preferences; list all valid keys; tmux settings |
Orienting in an Existing Workspace
Use app snapshot when you land in a session with existing panes and need to understand what's running without reading full screen contents for each pane:
cli-anything-iterm2 --json app snapshot
Returns name, current directory, foreground process, user.role label, and last visible output line for every session across all windows.
Naming convention — label panes when setting up a workspace so you can find them later:
cli-anything-iterm2 session set-var user.role "api-server"
cli-anything-iterm2 session set-var user.role "log-tail"
cli-anything-iterm2 session set-var user.role "editor"
app snapshot will surface these roles alongside process and path, giving you a full picture in one call.
Typical Agent Workflow
# 1. Orient — snapshot every session: name, path, process, role, last output line
cli-anything-iterm2 --json app snapshot
# 2. Establish context (saves window/tab/session IDs for subsequent commands)
cli-anything-iterm2 app current
# 3. Interact — no --session-id needed once context is set
cli-anything-iterm2 session send "git status"
cli-anything-iterm2 --json session scrollback --tail 200 --strip
# 4. Create a multi-pane workspace — label panes so snapshot identifies them later
cli-anything-iterm2 session split --vertical --use-as-context
cli-anything-iterm2 session send "python3 -m http.server 8000"
cli-anything-iterm2 session set-var user.role "http-server"
Reference Files
Read only what the task requires — each file is a single narrow concern (~10–30 lines):
| File | Read when you need... |
|---|---|
references/session-io.md |
Send text, inject bytes, read screen/scrollback, get selection |
references/session-control.md |
Split panes, activate/close sessions, resize, rename, session variables |
references/session-shell-integration.md |
wait-prompt, wait-command-end, get-prompt; reliable send→wait→read pattern |
references/layout-window-tab.md |
Create/close/resize windows and tabs, navigate split panes |
references/layout-arrangement.md |
Save and restore window layouts |
references/app-context.md |
Snapshot (orientation), status, context management, app vars, modal dialogs, file panels |
references/profile-pref.md |
Profiles list/get/presets, preferences read/write, tmux pref shortcuts |
references/broadcast-menu.md |
Broadcast keystrokes to multiple panes, invoke menu items |
references/tmux-commands.md |
All tmux CLI commands (bootstrap, send, tabs, create-window, set-visible) |
references/tmux-guide.md |
Full tmux -CC workflow, pane→session ID mapping |
references/json-session.md |
--json schemas for session, window, tab, screen, scrollback, inject |
references/json-tmux-app.md |
--json schemas for tmux, app dialogs, preferences, errors |
REPL Mode
Run without arguments for an interactive REPL that maintains context between commands:
cli-anything-iterm2
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
cli-anything-sts2
Command-line interface for Slay the Spire 2 - Control the real game through a local bridge mod HTTP API. Reads normalized game state and sends action commands for combat, navigation, rewards, and menu management.
musescore
CLI for music notation — transpose, export PDF/audio/MIDI, extract parts, manage instruments
cli-anything-videocaptioner
AI-powered video captioning — transcribe speech, optimize/translate subtitles, burn into video with beautiful customizable styles (ASS outline or rounded background). Free ASR and translation included.
cli-anything-openscreen
Command-line interface for Openscreen — a screen recording editor. A stateful CLI for editing screen recordings with zoom, speed ramps, trim, crop, annotations, and polished exports. Built on the Openscreen JSON project format with ffmpeg as the rendering backend. Designed for AI agents and power users who need programmatic video editing.
cli-anything-adguardhome
Command-line interface for AdGuard Home - Network-wide ad blocking and DNS management via AdGuard Home REST API. Designed for AI agents and power users who need to manage filtering, DNS rewrites, clients, DHCP, and query logs without a GUI.
cli-anything-intelwatch
Zero friction. Full context. Competitive intelligence, M&A due diligence, and OSINT directly from your terminal. Uses Node.js/npx.
Didn't find tool you were looking for?