gmail-policy

Enforce policy-aware intent checks before Gmail actions via gws CLI or MCP tools.

3|1|Updated Nov 10, 2025

Search Tags:#compliance #safety #gmail #policy #drafts #send #gws

Ready to run

npx skills add https://github.com/metraton/gaia --skill gmail-policy

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: gmail-policy
Source: https://github.com/metraton/gaia/tree/main/skills/gmail-policy
Command: npx skills add https://github.com/metraton/gaia --skill gmail-policy

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guards Gmail actions by ensuring intent is read and commands are classified correctly, preventing unsafe or unintended operations.

Core Features & Use Cases

Pre-action intent checks to distinguish between context and commands before any Gmail operation
Anti-Drift safety rules that default to drafting in sensitive contexts and provide clear action routing
Guidance for gws Gmail policy, including drafting and sending workflows, labeling, and triage across Gmail MCP tools
Use Case: review inbox and draft context-aware replies; or approve and send only after user confirmation

Quick Start

Use this skill to apply Gmail policy checks to a new or ongoing message before performing actions.

Frequently Asked Questions about gmail-policy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I enforce safety guardrails before sending Gmail messages via automation?▼

Gmail policy guardrails enforce safety by applying pre-action intent checks that distinguish user context from send commands, preventing unintended email operations. They route actions to draft or send based on sensitivity and user confirmation.

How does policy-driven Gmail triage handle sensitive data in message drafts?▼

Policy-driven Gmail triage handles sensitive data by applying anti-drift safety rules that default to drafting instead of sending. It evaluates user intent and triages sensitive content before any Gmail action is executed.

Can I use this Gmail policy skill with both gws CLI and Gmail MCP tools?▼

Yes, this Gmail policy skill works with both gws CLI and Gmail MCP tools. It applies frontmatter-driven controls and safety checks across these interfaces to govern drafting, sending, and labeling workflows.

What is the best way to prevent unintended Gmail send actions in automated workflows?▼

The best way to prevent unintended Gmail send actions is applying anti-drift safety rules that classify user intent and default to drafting. This policy-driven approach requires explicit confirmation before executing any send command.

Why does my Gmail automation draft messages instead of sending them directly?▼

Your Gmail automation drafts messages instead of sending because anti-drift safety rules default to drafting in sensitive contexts. The policy requires evaluating user intent and confirming before classifying an action as a send command.