TempMail OTP MCP Server
Enables AI agents to generate temporary email addresses, receive emails, and automatically extract OTP codes and links from incoming messages for automation and testing workflows.
README
TempMail OTP — Self-Hosted Temporary Email Generator
Receive-only temp email server with automatic OTP extraction for security research, QA automation, and disposable identity workflows.
Features
- SMTP catch-all — accepts all inbound email on your domain (port 25, no auth)
- Real-time inbox — SSE stream pushes new emails instantly to the browser
- OTP extraction — automatic detection of verification codes (4–8 digit, alphanumeric)
- Link extraction — parse all URLs from plain-text and HTML email bodies
- REST API — authenticated endpoints for email generation, reading, OTP, and link extraction
- MCP integration — stdio-based Model Context Protocol server for AI agents (Hermes, Claude, etc.)
- Docker — multi-stage build, SQLite persistence, health check included
- Dark/light mode — built with Tailwind CSS + shadcn/ui
- DNS verification — built-in
/api/dns-checkendpoint validates MX, SPF, DMARC records - Multi-key support — manage API keys via the web UI or
/api/keysendpoint
Prerequisites
- Node.js 20+ (with
tsxfor MCP;npmfor package management) - Domain name — configured with MX, SPF, and DMARC records
- Port 25 access — required for SMTP receiving; unavailable on most home ISPs. Use a VPS (Oracle Cloud Free Tier, AWS EC2, DigitalOcean) or port forwarding if supported
Quick Start
1. Clone and install
git clone <repo-url> website-email
cd website-email
npm install
2. Configure environment
cp .env.example .env
Edit .env with your domain and a strong API key:
DOMAIN=mail.yourdomain.com
API_KEY=your-generated-api-key
Generate a secure API key:
openssl rand -hex 32
3. Set up DNS records
Configure MX, SPF, and DMARC records for your domain. See the complete guide:
Quick verification after DNS propagates:
dig MX yourdomain.com +short
dig TXT yourdomain.com +short | grep spf
dig TXT _dmarc.yourdomain.com +short
4. Start the servers
Terminal 1 — Next.js web UI + API:
npm run dev
Terminal 2 — SMTP server (receives mail):
npx tsx src/infrastructure/smtp/start.ts
5. Open the app
http://localhost:3000
Generate a temp email address, send a test email to it, and watch it appear in real-time.
Docker Deployment
1. Configure .env
Same as Quick Start — fill in DOMAIN and API_KEY.
2. Build and start
docker compose up -d
3. Verify
curl http://localhost:3000/api/config
# {"success":true,"data":{"domain":"mail.yourdomain.com"}}
Note: Docker Compose exposes port 3000 only (web UI + API). Run the SMTP server directly on the host for port 25:
npx tsx src/infrastructure/smtp/start.ts
For production, use a reverse proxy (nginx/Caddy) for TLS termination and run SMTP via a process manager (systemd, pm2).
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
DOMAIN |
Yes | — | Your domain for receiving email (e.g. mail.example.com) |
API_KEY |
Yes | — | Secret key for API and MCP authentication |
SMTP_PORT |
No | 25 |
Port for inbound SMTP |
SMTP_HOST |
No | 0.0.0.0 |
Bind address for SMTP server |
DB_PATH |
No | ./data/emails.db |
SQLite database file path |
NEXT_PUBLIC_APP_URL |
No | http://localhost:3000 |
Public URL of the application |
REST API
All endpoints (except /api/config and /api/dns-check) require an x-api-key header. Response format: { "success": true, "data": ... } or { "success": false, "error": "..." }.
Email Addresses
| Method | Endpoint | Description |
|---|---|---|
GET |
/api/addresses |
List all addresses |
POST |
/api/addresses |
Generate new address (optional body: { "address": "custom@domain" }) |
DELETE |
/api/addresses/:id |
Delete an address and its emails |
Emails
| Method | Endpoint | Description |
|---|---|---|
POST |
/api/emails/generate |
Generate address (shorthand) |
GET |
/api/emails/:address |
List emails for an address |
GET |
/api/emails/:address/:id |
Read a specific email |
DELETE |
/api/emails/:address/:id |
Delete a specific email |
POST |
/api/emails/:address/:id/otp |
Extract OTP from email |
GET |
/api/emails/:address/:id/links |
Extract links from email |
GET |
/api/emails/:address/stream |
SSE stream for real-time email events |
Misc
| Method | Endpoint | Description |
|---|---|---|
GET |
/api/config |
Get server domain (no auth) |
GET |
/api/dns-check |
Verify DNS records (MX, SPF, DMARC) |
GET |
/api/keys |
List API keys |
POST |
/api/keys |
Create new API key |
Example: Full workflow
KEY="your-api-key"
DOMAIN="mail.example.com"
# Generate an address
curl -sH "x-api-key: $KEY" -X POST http://localhost:3000/api/emails/generate | jq
# List emails
curl -sH "x-api-key: $KEY" "http://localhost:3000/api/emails/temp-abc123@$DOMAIN" | jq
# Read email and extract OTP
curl -sH "x-api-key: $KEY" -X POST \
"http://localhost:3000/api/emails/temp-abc123@$DOMAIN/eml_xxx/otp" | jq
MCP Integration
The MCP server exposes 5 tools for AI agents via stdio transport:
| Tool | Description |
|---|---|
temp_email_generate |
Generate a new temporary email address |
temp_email_list |
List emails for an address |
temp_email_read |
Read a specific email by ID |
temp_email_extract_otp |
Extract OTP/verification code from an email |
temp_email_extract_links |
Extract all links from an email |
Configuration and workflow documentation:
→ docs/hermes-integration.md — Hermes MCP configuration → docs/chrome-devtools-mcp.md — Browser automation with Chrome DevTools MCP → docs/workflow-examples.md — Copy-paste ready prompts
Quick test:
DOMAIN=mail.example.com API_KEY=your-key npx tsx src/infrastructure/mcp/start.ts
Architecture
┌─────────────────────────────────────────────┐
│ Internet │
│ (inbound SMTP on port 25 from any server) │
└──────────────────┬──────────────────────────┘
│
▼
┌──────────────────────────────────────────────┐
│ SMTP Server (smtp-server, port 25) │
│ - Accepts all mail for configured DOMAIN │
│ - Parses MIME with mailparser │
│ - Extracts OTP codes automatically │
│ - Stores to SQLite │
└──────────────────┬───────────────────────────┘
│
▼
┌──────────────────────────────────────────────┐
│ SQLite (better-sqlite3, WAL mode) │
│ - addresses, emails, api_keys tables │
│ - Event emitter for SSE notifications │
└────────┬──────────────────┬──────────────────┘
│ │
▼ ▼
┌─────────────────┐ ┌─────────────────────────┐
│ Next.js App │ │ MCP Server (stdio) │
│ - Web UI │ │ - 5 tools for AI agents │
│ - REST API │ │ - Direct DB access │
│ - SSE stream │ │ │
└─────────────────┘ └─────────────────────────┘
Stack: Next.js 14 + TypeScript + Tailwind CSS + shadcn/ui + better-sqlite3 + smtp-server + mailparser
Additional Documentation
- docs/dns-setup.md — DNS record configuration (MX, SPF, DMARC)
- docs/deployment.md — Production deployment (Oracle Cloud, SSL)
- docs/troubleshooting.md — Common issues and solutions
- docs/hermes-integration.md — Hermes AI agent MCP configuration
- docs/chrome-devtools-mcp.md — Browser automation workflows
- docs/workflow-examples.md — End-to-end automation examples
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.