Gmail Skill for AI Agents for Openclaw

A professional-grade integration allowing AI agents to read, send, and manage Gmail threads and labels using the official Google API.

terrycarter1985
v1.0.0
May 12, 2026
0
660
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install tc-gmail

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install tc-gmail 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 Gmail Skill for AI Agents?

The Gmail skill is a robust communication interface designed for modern AI agents within the Openclaw Skills ecosystem. By leveraging the official Gmail API and OAuth 2.0 authentication, it enables AI models to act as sophisticated personal assistants capable of handling high-volume email workflows with precision.

This skill goes beyond simple message sending; it allows for deep inspection of email threads, attachment handling, and complex organizational tasks through label management. It is an essential component for developers building autonomous agents that need to interact with the world via the most ubiquitous communication protocol: email.

Gmail Skill for AI Agents Use Cases

  • Automatically drafting and sending follow-up emails based on meeting notes.
  • Searching and extracting data from invoices or receipts using specific date filters.
  • Organizing overflowing inboxes by automatically applying stars or custom labels to urgent messages.
  • Building automated customer support triaging systems that flag unread messages for human review.

How Gmail Skill for AI Agents Works

  1. The AI agent authenticates with the Google Cloud Platform using provided OAuth 2.0 credentials.
  2. The agent executes a tool call (e.g., gmail_search) to locate specific threads based on a query.
  3. Content is retrieved and parsed into a structured format for the LLM to process.
  4. Based on its reasoning, the agent can then send replies, modify labels, or manage drafts to complete the task cycle.

Gmail Skill for AI Agents Setup

To integrate this into your environment, ensure you have the required Python dependencies and set your environment variables:

pip install google-api-python-client google-auth-httplib2 google-auth-oauthlib

Configure your environment variables in your .env file or shell:

export GMAIL_CLIENT_ID="your_google_client_id"
export GMAIL_CLIENT_SECRET="your_google_client_secret"
export GMAIL_REFRESH_TOKEN="your_oauth_refresh_token"

Gmail Skill for AI Agents Data Schema & Taxonomy

The skill interacts with the Gmail API's native data structures. Below is how common metadata and search operations are organized:

Feature Description
Query Operators Supports standard filters like is:unread, from:, and has:attachment.
Message ID Unique hex strings (e.g., msg_12345abcdef) used for direct retrieval.
Labels System labels (STARRED, UNREAD, INBOX) and user-defined custom labels.
Formats Options for minimal, metadata, or full message body retrieval.

Gmail Skill for AI Agents Advanced Features

  • Multi-recipient support including CC and BCC fields for formal correspondence.
  • Support for file attachments by passing workspace paths directly to the sending tool.
  • Advanced filtering using Gmail's complex search syntax (before/after dates and boolean logic).
  • Granular label modification allowing simultaneous adding and removing of tags for state management.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*