beeper-api-cli

What This Skill Does

The beeper-api-cli skill serves as a robust LLM-friendly interface for the Beeper messaging platform. It allows OpenClaw to interact with your cross-platform chat accounts—including WhatsApp, Telegram, Signal, Instagram, Twitter/X, LinkedIn, and Facebook Messenger—directly from the command line. By wrapping the Beeper local API, this skill enables your AI agent to read messages, manage chat threads, and send replies across these disparate services from a unified hub.

Installation

To install, run: clawhub install openclaw/skills/skills/nerveband/beeper-api-cli. After installation, ensure Beeper Desktop is running. Go to Settings > Advanced > API in the Beeper app to enable API access and generate your Bearer token. Set the token as an environment variable in your shell: export BEEPER_TOKEN="your-token-here". By default, the skill connects to http://[::1]:23373 on localhost, but this can be overridden via the BEEPER_API_URL variable if connecting across a network.

Use Cases

• Unified Inbox Management: Let your AI summarize unread messages from multiple platforms like WhatsApp and Telegram in a single report.
• Smart Reply Drafting: Have the AI draft professional responses based on chat history, while maintaining strict adherence to human-in-the-loop verification.
• Automated Alerts: Use the agent to send notifications or status updates via your preferred messaging apps using simple natural language commands.
• Efficient Thread Retrieval: Quickly find specific conversations or contacts across your entire chat ecosystem without switching apps.

Example Prompts

1. "OpenClaw, please list my unread messages from WhatsApp and summarize the top three urgent ones."
2. "Draft a polite reply to Sarah on Telegram acknowledging her project update and suggest a meeting for Tuesday at 10 AM. Show me the draft first."
3. "Show me the last 5 messages from the group chat named 'Team Alpha' to get context on our current task."

Tips & Limitations

CRITICAL SAFETY NOTE: The agent is strictly programmed to never send messages without your explicit, verbal approval. Even when an AI draft is generated, it will pause and display the full text for your review. Always ensure Beeper Desktop remains open, as the CLI acts as a bridge to the locally running instance. If the CLI returns an 'Unauthorized' error, double-check your BEEPER_TOKEN and ensure the Beeper Desktop API toggle is still enabled. For remote execution, remember to add your machine's IP to the Beeper API settings.