postwall
v1.2.0
Published
CLI for PostWall email security gateway - AI agent email with human approval
Maintainers
Readme
PostWall CLI
A security layer between AI agents and email. Agents can only read emails that have been approved by a human, and can only send emails after human review.
Installation
npm install -g postwallQuick Start
Get your API key from the PostWall Dashboard
Authenticate:
postwall auth pw_your_api_key_here- Check for approved emails:
postwall checkCommands
| Command | Description |
|---------|-------------|
| postwall auth <key> | Authenticate with your API key |
| postwall check | Get count of unread approved emails |
| postwall inbox | List unread approved emails |
| postwall read <id> | Read a specific email (marks as read) |
| postwall draft --to --subject --body | Submit email draft for approval |
| postwall status <id> | Check draft status |
| postwall drafts | List all drafts |
All commands support --json flag for structured output.
How It Works
- Inbound emails: Emails arrive in your Gmail → You approve them in PostWall dashboard → Agent can read them via CLI
- Outbound emails: Agent submits draft via CLI → You review in dashboard → Approve to send
OpenClaw/ClawdBot Integration
This CLI is designed to work as an OpenClaw skill. After installing via npm, link to your skills directory:
npm install -g postwall
ln -s $(npm root -g)/postwall ~/.openclaw/skills/postwallLinks
License
MIT
