A unified messaging skill for Clawdbot that integrates with the Beeper Desktop API to manage chats across multiple platforms from a single interface.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install claw-me-maybe
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install claw-me-maybe using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is claw-me-maybe?
claw-me-maybe is a comprehensive integration designed to connect your AI coding agent to the Beeper ecosystem. By leveraging the Beeper Desktop API, this skill allows your agent to interact with over a dozen messaging platforms—including WhatsApp, Telegram, Signal, Discord, Slack, and iMessage—through a single unified interface. This is a standout example of how Openclaw Skills can centralize communication workflows, enabling agents to search history, send messages, and manage notifications across fragmented networks without leaving the development environment.
The skill operates locally and privately, ensuring that your message data remains on your machine while providing the agent with the ability to perform complex tasks like summarizing unread threads or setting reminders for specific chats. It bridges the gap between your technical workspace and your personal or professional communication channels.
claw-me-maybe Use Cases
- Searching through all connected chat platforms for specific information or past conversations.
- Sending messages to contacts on WhatsApp, Slack, or Telegram via simple natural language commands.
- Generating summaries of unread messages across all networks to stay updated quickly.
- Setting follow-up reminders for important threads directly through the agent.
- Automating inbox maintenance by marking conversations as read or reacting to messages with emojis.
How claw-me-maybe Works
- The skill connects to the Beeper Desktop API running on the user's local machine (typically at localhost:23373).
- Upon invocation, the agent authenticates using a Beeper Access Token provided in the configuration.
- The agent retrieves a list of active accounts and chats across all bridged services (e.g., Signal, Discord).
- When a command is given, the skill sends localized HTTP requests to the Beeper API to perform actions like searching message history or posting new text.
- Beeper handles the protocol-specific communication with the external messaging networks and returns the results to the agent.
claw-me-maybe Setup
First, ensure Beeper Desktop is installed and running. Enable the API by navigating to Settings > Developers and toggling Beeper Desktop API to ON. To configure the skill for your agent, add your access token to your configuration file:
# Edit ~/.clawdbot/clawdbot.json
{
"skills": {
"entries": {
"claw-me-maybe": {
"enabled": true,
"env": {
"BEEPER_ACCESS_TOKEN": "your-token-here"
}
}
}
}
}
You can verify the connection by running curl http://localhost:23373/health in your terminal.
claw-me-maybe Data Schema & Taxonomy
The skill interacts with several key data entities provided by the Beeper API:
| Entity | Data Description |
|---|---|
| Accounts | List of connected platforms including service type (whatsapp, slack, etc.) and connection status. |
| Chats | Contains chat IDs, display names, unread counts, and metadata for individual conversations. |
| Messages | Includes message text, sender details, timestamps, and reaction arrays. |
| Contacts | Stores contact names, phone numbers, and avatars across different networks. |
| Reminders | Metadata for scheduled chat follow-ups and alerts. |
claw-me-maybe Advanced Features
- Multi-platform support for over 12 networks including iMessage (macOS) and Google Messages.
- Automated unread summaries that aggregate notifications from all active bridges.
- Attachment handling for downloading files, images, and media directly from chat threads.
- Batch processing capabilities, such as marking all unread messages as read across the entire inbox.
- Local-first architecture ensuring that sensitive messaging data is never sent to third-party servers through the skill.
SKILL.md
Loading
Related Openclaw Skills
Fundamental Stock Analysis
A structured equity analysis tool that scores and ranks stocks using a comprehensive fundamental playbook.
Robonet Workbench
A comprehensive MCP-powered toolkit for building, optimizing, and deploying automated trading strategies across crypto and prediction markets.
CounterClaw
CounterClaw is a production-grade defensive interceptor designed to protect AI agents from malicious prompt injections and sensitive data leaks.
Paperless-NGX CLI for AI Agents
A comprehensive CLI integration for Paperless-NGX that allows AI agents to search, retrieve, and organize digital documents using the ppls tool.
Pinch to Post (WP Pinch)
A comprehensive suite of 54 MCP tools to manage WordPress content, media, and site health directly from your AI agent.
FreshRSS
A powerful bridge between your self-hosted FreshRSS instance and your AI agent for streamlined news consumption.
Featured*
