A secure, multi-account command-line email client designed for AI agents using Openclaw Skills to safely read, search, and manage Gmail, Outlook, and Exchange inboxes.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install email-gmail-outlook
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 email-gmail-outlook 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 Email Management (Gmail, Outlook & Exchange)?
The porteden email skill is a high-performance, developer-focused command-line tool designed to integrate seamless, secure email capabilities into AI workflows. Built as a secure alternative to complex API integrations, this skill allows agents to search, retrieve, send, reply to, and archive emails across Gmail, Outlook, and Exchange directories with multi-account support. By leveraging compact JSON outputs, the tool maximizes efficiency and significantly reduces context window usage for LLM automation.
Integrating this tool with Openclaw Skills empowers teams to build robust, secure, and production-ready email automation agents. Built-in security boundaries prevent unsafe mutations, isolate enterprise accounts via profiles, and natively secure credentials using system keyrings.
Email Management (Gmail, Outlook & Exchange) Use Cases
- Automating inbox triaging and categorization using AI agents to classify incoming emails based on intent or priority.
- Building smart draft generators that construct context-aware replies to customer inquiries or support tickets.
- Executing semantic searches across entire conversation histories to synthesize context for software development or client management.
- Developing scheduled notification systems that aggregate reports and forward them directly to specific target recipients.
- Integrating multi-account productivity workflows that isolate corporate Exchange directories from personal Gmail accounts safely.
How Email Management (Gmail, Outlook & Exchange) Works
- Authentication & Session Init: The CLI authenticates with providers via browser-based OAuth or direct API keys, safely storing credentials in the local system keyring.
- Query & Search Formulation: The AI agent issues structured commands using date ranges, labels, or custom keywords to query the active mail server.
- Token-Optimized Fetching: The tool retrieves the messages, applying the AI-optimized compact flag (-jc) to minify body previews and eliminate unnecessary token overhead.
- Safety Evaluation: For destructive or mutating actions (like sending, replying, or deleting), the agent prompts for human confirmation to prevent accidental transmissions.
- Provider Execution: Once authorized, the command is executed natively through the provider's API using provider-prefixed message identifiers.
Email Management (Gmail, Outlook & Exchange) Setup
Installation
Install the porteden CLI using Homebrew or Go:
# Install via Homebrew
brew install porteden/tap/porteden
# Alternatively, install via Go
go install github.com/porteden/cli/cmd/porteden@latest
Authentication
Log in once to store credentials securely in your system's keyring:
# Browser-based login (recommended)
porteden auth login
# Alternatively, log in directly using a token
porteden auth login --token <your_key>
# Verify authentication status
porteden auth status
You can also configure the environment directly by setting the PE_API_KEY environment variable.
Email Management (Gmail, Outlook & Exchange) Data Schema & Taxonomy
Environment Variables
| Variable | Description |
|---|---|
PE_API_KEY |
API access key for seamless non-interactive CLI authorization. |
PE_PROFILE |
Sets the default mail profile (e.g., work, personal) to avoid manual selection. |
PE_TIMEZONE |
Configures the timezone context for time-filtered queries. |
PE_FORMAT |
Forces specific output serialization formats (e.g., json). |
Object Schema Format
The tool returns structured, token-saving compact JSON payloads when invoked with the -jc flag (JSON Compact):
- Provider-Prefixed IDs: All identifiers are unified and prefixed based on source (e.g.,
google:abc123orm365:xyz789). - Compact Payload (
-jc): Automatically truncates body previews, drops extensive attachment details, and filters redundant metadata to minimize LLM token usage. - Pagination Metadata: Large message queries return pagination markers like
hasMoreandnextPageTokenkeys inside the parent JSON envelope.
Email Management (Gmail, Outlook & Exchange) Advanced Features
- Multi-Account Profiles: Use the
--profileflag orPE_PROFILEenvironment variable to isolate different Gmail and Outlook directories natively. - AI-Optimized JSON Formatting: Compact JSON serialization (
-jc) dramatically compresses body previews and removes heavy attachments to keep agent prompt contexts clean. - Granular Least-Privilege Scopes: Configure narrow provider-level authorization scopes during authentication to satisfy strict enterprise security policies.
- Auto-Pagination: Fetch complete histories by passing the
--allflag, allowing the tool to recursively navigate page tokens. - Smart Keyring Management: Fully supports native platform keyrings for seamless credential caching and fast revocation with
porteden auth logout.
SKILL.md
Loading
Related Openclaw Skills
MCP Token Auditor
Prevent subagent crashes caused by tool schema bloat by auditing your MCP configuration and generating optimized, role-specific setups.
A-Share Investment Meeting Preparer (Meeting Preparer CN)
An AI-driven research agent that automatically compiles regulatory filings, analyst consensus, and financial metrics to produce professional briefing packages for A-share investor meetings.
WeChat Mini Program to uni-app Converter (Vue3 & TypeScript)
An automated migration skill that translates native WeChat Mini Programs into cross-platform uni-app projects featuring Vue3 and TypeScript.
Subagent Health Monitor
A real-time MCP server for Claude Code that monitors subagent health, minimizes token waste, and prevents silent agent failures.
Smart Model Router
An intelligent, intent-based model router that automatically detects your prompt's context and routes it to the optimal AI agent.
Local MCP Server
A privacy-first, lightweight Model Context Protocol server for Termux and Linux environments that exposes file access, grep, and bash execution tools to offline Ollama models.
Featured*
