A comprehensive Gmail API integration for AI agents that provides managed OAuth authentication to read, send, and organize emails and threads.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install gmail-1-0-6
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install gmail-1-0-6 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Gmail skill is a specialized integration designed to bridge the gap between AI coding agents and the Google Mail ecosystem. By leveraging a managed OAuth gateway, this tool allows Openclaw Skills to perform complex email operations—such as retrieving message threads, modifying labels, and sending drafts—without requiring the developer to manage complex token refresh logic manually. It serves as a critical component for building sophisticated communication workflows within the Openclaw Skills framework.
Built with developer productivity in mind, this skill abstracts the native Google REST API into a predictable gateway. Whether you are building an automated support agent or a personal assistant, the Gmail integration ensures that your AI has secure, authenticated access to mailbox data while maintaining high performance and reliability.
To get started with this skill in your Openclaw Skills environment, follow these steps:
export MATON_API_KEY="YOUR_API_KEY"
# Example using Python to create a connection
python <<'EOF'
import urllib.request, os, json
data = json.dumps({'app': 'google-mail'}).encode()
req = urllib.request.Request('https://ctrl.maton.ai/connections', data=data, method='POST')
req.add_header('Authorization', f'Bearer {os.environ["MATON_API_KEY"]}')
req.add_header('Content-Type', 'application/json')
print(json.dumps(json.load(urllib.request.urlopen(req)), indent=2))
EOF
The Gmail skill utilizes standard Google Mail resource structures. Below is how the data is organized for Openclaw Skills users:
| Resource | Description | Key Fields |
|---|---|---|
| Messages | Individual email records | id, threadId, labelIds, snippet, raw |
| Threads | Grouped conversations | id, historyId, messages |
| Labels | Inbox organization tags | id, name, type, messagesTotal |
| Drafts | Unsent email messages | id, message |
All raw email content is base64url encoded to ensure safe transmission of multi-part data.
Maton-Connection header to switch between different Gmail accounts within the same Openclaw Skills workflow.is:unread, has:attachment, and after:YYYY/MM/DD.Loading
An autonomous, spec-driven development framework that uses the OpenSpec CLI and Claude Code to orchestrate the full software lifecycle from issue to pull request.

A robust API-driven charting tool that generates professional PNG visualizations from raw data via simple CLI commands or JSON configurations.

A technical optimization skill designed to minimize Feishu API consumption by extending probe cache durations in OpenClaw.

A professional-grade A-share market report generator that produces detailed HTML and image-based technical analysis for personal research.

A specialized tool for tuning and calibrating Megasquirt engine management systems via TunerStudio and automated MSQ file analysis.

A centralized registry and verification platform for discovering and listing autonomous AI agents.








































