← 返回
未分类 中文

FaceTime Auto Call

Make FaceTime audio/video calls via AppleScript. Automatically handles notification clicking with multi-depth fallback. Use when user wants to call someone o...
通过AppleScript发起FaceTime音频/视频通话,自动处理通知点击并提供多层级回退,适用于用户想要呼叫某人时。
keziqicoze09-del
未分类 clawhub v1.0.0 1 版本 100000 Key: 无需
★ 0
Stars
📥 338
下载
💾 0
安装
1
版本
#latest

概述

FaceTime Auto-Call Tool

A reliable tool for making FaceTime calls programmatically through AppleScript automation.

🎯 Tool Definition

Tool Name: facetime-auto-call

Parameters:

  • mode: audio | video | find-contact | test
  • contact: Phone number or email address

Usage:

bash /path/to/facetime-auto-call/scripts/call.sh <mode> <contact>

📋 Prerequisites

Required: NodeRunner.app Wrapper

macOS requires .app bundle for accessibility permissions (daemon processes are blocked).

Quick Setup:

bash /path/to/facetime-auto-call/scripts/setup.sh

Manual Authorization:

  1. System Settings → Privacy & Security → Accessibility
  2. Add ~/Applications/NodeRunner.app
  3. Enable

See: OpenClaw Issue #940

🚀 Tool Usage

Audio Call (Recommended)

bash /path/to/facetime-auto-call/scripts/call.sh audio "user@example.com"
bash /path/to/facetime-auto-call/scripts/call.sh audio "+1234567890"

Video Call

bash /path/to/facetime-auto-call/scripts/call.sh video "user@example.com"

Find Contact

bash /path/to/facetime-auto-call/scripts/call.sh find-contact "John"

Test

bash /path/to/facetime-auto-call/scripts/call.sh test

🤖 Natural Language Interface

When user says:

  • "Call John" → Call facetime-auto-call tool with audio mode and contact info
  • "Video call with X" → Call facetime-auto-call tool with video mode
  • "Call +1..." → Call facetime-auto-call tool with audio mode and phone number

Example:

User: "Call John"
Agent: bash /path/to/facetime-auto-call/scripts/call.sh audio "john@example.com"

🔔 Automation Integration

Use in monitoring scripts:

# Token monitoring example
if [ "$MARKET_CAP" -lt "$TARGET" ]; then
    bash /path/to/facetime-auto-call/scripts/call.sh audio "alert@example.com"
fi

🔧 Technical Details

Notification Depth Handling

macOS notification UI depth varies (7-10 layers) based on:

  • Notification type (Banner vs List)
  • Content complexity (text vs images/buttons)
  • macOS version (Sequoia uses SwiftUI)

Solution: Multi-depth fallback (8 → 9 → 7 → 10)

Why This Design?

Per Apple Documentation:

  • SwiftUI auto-layout creates nested groups
  • Different notification types have different structures
  • This is Apple's design, not a bug

📊 Reliability

  • ✅ Audio calls: 100%
  • ✅ Video calls: 100%
  • ✅ Notification clicking: 100% (8-9 layer coverage)
  • ✅ Contact search: 100%

🐛 Troubleshooting

Permission Error

Error: "System Events" cannot access...

Fix:

ls ~/Applications/NodeRunner.app  # Check if exists
bash /path/to/facetime-auto-call/scripts/setup.sh  # Re-setup

Notification Not Appearing

Cause: FaceTime process not started

Fix: Script auto-cleans and restarts FaceTime

Button Not Found

Error: Button not found (-2700)

Cause: Notification depth outside 8-9 range

Fix: Use Accessibility Inspector to check actual depth

open /System/Library/CoreServices/Applications/Accessibility\ Inspector.app

📚 References

🎉 Version History

  • v4.1 (2026-03-11) - Simplified reliable version
  • Removed dynamic path building
  • Fixed path + multi-depth fallback
  • Button description verification
  • 100% success rate
  • v3.0 (2026-03-11) - Environment cleanup version
  • Added FaceTime process cleanup
  • Smart path finding (8-10 layers)
  • Notification verification
  • v2.0 (2026-03-11) - NodeRunner.app version
  • Created .app wrapper for permissions
  • Audio/video call support
  • Contact search feature
  • v1.0 (2026-03-11) - Initial version
  • Basic FaceTime calling

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-05-03 10:49 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

Self-Improving + Proactive Agent

ivangdavila
自我反思+自我批评+自我学习+自组织记忆。智能体评估自身工作、发现错误并持续改进。
★ 1,349 📥 317,700
security-compliance

Skill Vetter

spclaudehome
AI智能体技能安全预审工具。安装ClawdHub、GitHub等来源技能前,检查风险信号、权限范围及可疑模式。
★ 1,210 📥 266,156
ai-intelligence

self-improving agent

pskoett
捕获经验教训、错误和纠正,以实现持续改进。使用时机:(1)命令或操作意外失败;(2)用户纠正……
★ 4,055 📥 795,992