Forward any email you receive that you think might be suspicious to sus@inbound.delivery and within a couple of minutes you will receive an AI-powered security review analyzing whether the email is safe or potentially malicious.
This is a Next.js project built with:
- Next.js 16
- Bun as the package manager
- Vercel Workflow for async email processing
- OpenAI GPT-5 for spam detection and reply generation
- Inbound Email API for receiving and sending emails
First, install dependencies:
bun installThen, run the development server:
bun devOpen http://localhost:3000 with your browser to see the result.
susbound/
├── app/
│ ├── api/
│ │ └── inbound/
│ │ └── route.ts # Webhook endpoint that receives emails
│ ├── page.tsx # Home page
│ └── layout.tsx # Root layout
├── workflows/
│ └── scanAndReply.ts # Workflow that processes emails:
│ # 1. Analyzes email for spam
│ # 2. Generates AI reply
│ # 3. Sends reply via Inbound API
└── next.config.ts # Next.js config with Workflow integration
Required environment variables:
INBOUND_API_KEY- Your Inbound API key for sending/receiving emailsOPENAI_API_KEY- OpenAI API key for GPT-5 modelINBOUND_REPLY_FROM- Verified email address to send replies from (optional, defaults to susdev@inbound.delivery)
- User forwards suspicious email to
sus@inbound.delivery - Inbound sends webhook to
/api/inboundendpoint - API route starts async workflow
scanAndReply - Workflow performs three steps:
- Step 1: Analyzes email content and headers for spam/malicious indicators using GPT-5
- Step 2: Generates a concise safety evaluation reply
- Step 3: Sends the reply back to the user via Inbound API
powered by
