Claude Code Control

# Claude Code Control

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "Claude Code Control" with this command: npx skills add melichar-m/claude-code-control

Claude Code Control

Control Claude Code programmatically through visible Terminal.app windows on macOS.

How It Works

Uses AppleScript to:

  1. Open Terminal.app and launch claude code in a project directory
  2. Type commands via System Events keystrokes
  3. Capture screenshots of just the Terminal window (not full screen)
  4. Record full sessions with timestamped logs

Requirements

  • macOS
  • Node.js 18+
  • Claude Code installed and authenticated
  • Accessibility permissions for Terminal.app + Script Editor (System Settings → Privacy & Security → Accessibility)

Usage

const cc = require('./index');

// Launch Claude Code visibly
const session = await cc.launch('/path/to/project');

// Send a command (types it + presses Enter)
const result = await cc.send(session, 'write tests for app.py', 30);
// result.screenshot → path to Terminal window screenshot

// Save session recording
await cc.saveSession(session, './recording.json');

// Close
await cc.close(session);

API

FunctionDescription
launch(path, opts?)Open Terminal + start Claude Code. Returns session ID
send(id, command, waitSec?)Type command, wait, screenshot. Returns {screenshot, duration_ms}
verifyScreen(id, desc)Take a verification screenshot
approveSecurity(id)Handle "trust this folder" prompt
handleLogin(id)Send /login command
saveSession(id, path)Save session log to JSON
close(id) / closeAll()Exit Claude Code gracefully
takeScreenshot(path?)Capture Terminal window
focusTerminal()Bring Terminal to front

Pro Features (Coming Soon)

  • 🎬 Video recording of sessions
  • 🤖 Multi-agent / multi-terminal orchestration
  • 📊 Session analytics
  • 🔄 Session replay
  • 🌐 Remote control via SSH

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

Cortex Engine

Persistent cognitive memory for AI agents — query, record, review, and consolidate knowledge across sessions with spreading activation, FSRS scheduling, and...

Registry SourceRecently Updated
Coding

AI Image & Video Toolkit — Free Upscale, Face Enhance, BG Remove & Generation

Free local AI image and video processing toolkit with cloud AI generation. Local tools: upscale (Real-ESRGAN), face enhance (GFPGAN/CodeFormer), background r...

Registry SourceRecently Updated
Coding

agent-bom compliance

AI compliance and policy engine — evaluate scan results against OWASP LLM Top 10, MITRE ATLAS, EU AI Act, NIST AI RMF, and custom policy-as-code rules. Gener...

Registry SourceRecently Updated