A command-line interface for sending WhatsApp messages, searching history, and synchronizing chat data via AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install whats
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 whats 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 wacli?
wacli is a powerful command-line utility designed to bridge the gap between AI agents and the WhatsApp messaging ecosystem. As part of the Openclaw Skills library, it allows developers and AI models to perform outbound communication, manage group messages, and retrieve historical data without manual interaction. Unlike standard chat interfaces, this skill is specifically engineered for programmatic tasks like contacting third parties or performing deep searches within message archives.
The skill provides a secure way to handle WhatsApp data by storing information locally and requiring explicit confirmation for any message sent. By leveraging this tool within Openclaw Skills, users can automate repetitive communication workflows while maintaining full control over their messaging identity and data privacy.
wacli Use Cases
- Sending automated notifications or follow-ups to external WhatsApp contacts.
- Retrieving historical invoices, dates, or specific mentions from past conversations using keyword searches.
- Programmatically sharing documents and files to WhatsApp groups or individual Jabber IDs (JIDs).
- Syncing and backfilling chat history to provide AI agents with a comprehensive context for better decision-making.
- Managing high-volume messaging tasks through a developer-friendly CLI integrated into Openclaw Skills.
How wacli Works
- The user authenticates the CLI by running the auth command and scanning a QR code with their mobile device to establish a secure session.
- The skill identifies recipients by querying the chat list to find specific JIDs for individuals or group chats.
- When a messaging task is triggered, the AI agent uses the send command with required parameters for the recipient and the message body.
- For data retrieval, the skill executes search queries against the local message store, filtering by date or keyword.
- Continuous synchronization can be enabled to ensure the local database remains updated with the latest incoming and outgoing messages within the Openclaw Skills framework.
wacli Setup
To integrate this skill into your Openclaw Skills environment, first install the wacli binary using one of the following methods:
# Install via Homebrew
brew tap steipete/tap
brew install wacli
# Install via Go
go install github.com/steipete/wacli/cmd/wacli@latest
Once installed, initialize your account and verify the installation:
wacli auth
wacli doctor
wacli Data Schema & Taxonomy
The wacli skill organizes data in a structured local directory, typically located at ~/.wacli. It uses a consistent taxonomy for identifying contacts and messages:
| Component | Identification Format | Description |
|---|---|---|
| Individual Chats | [number]@s.whatsapp.net | Unique JID for direct messaging |
| Group Chats | [id]@g.us | Unique JID for group communication |
| Message Store | JSON / Local DB | Searchable history of text and media metadata |
| File Attachments | Local Path | References to files sent or received via the CLI |
All outputs can be rendered in machine-readable JSON to facilitate seamless parsing by other Openclaw Skills.
wacli Advanced Features
- Real-time background synchronization using the sync follow command to capture live message streams.
- Advanced history backfilling that allows the retrieval of older messages by making specific requests to the WhatsApp server.
- Granular search capabilities allowing for date-range filtering and specific chat targeting.
- Support for machine-readable output across all commands to enable complex multi-agent automation within Openclaw Skills.
- Capability to send various file types with custom captions directly from the command line.
SKILL.md
Loading
Related Openclaw Skills
Spotify Player
A terminal-based Spotify controller that allows AI agents and developers to manage music playback and searches via command-line interfaces.
Xian SDK for Python
A comprehensive Python SDK for building decentralized applications, managing cryptographic wallets, and interacting with smart contracts on the Xian blockchain.
Xian Node Management
A professional toolkit for deploying, configuring, and managing Xian blockchain nodes using Docker and CometBFT.
Three Minds Multi-Agent Collaboration System
A multi-agent orchestration framework that launches three distinct AI personas to collaborate, refactor, and review code until a unanimous consensus is reached.
ConvertAgent
A unified command-line interface for reliable file format conversions across documents, images, spreadsheets, and media.
Xiaohongshu Long-Form Publisher
An automated workflow for publishing professional long-form articles on the Xiaohongshu creator platform via the built-in browser.
Featured*
