AtData SafeToSend MCP Server
@avivshafir
atdata-email-verification-mcp-server
Overview
What is AtData SafeToSend MCP Server?
A Model Context Protocol (MCP) server that provides email verification services using AtData’s SafeToSend API. It allows you to verify individual or batch email addresses to filter out invalid and high-risk ones, resulting in higher open rates, clicks, and conversions. Built for developers and AI assistants that need reliable email validation.
How to use AtData SafeToSend MCP Server?
Clone the repository, install dependencies with uv or pip, and set the ATDATA_API_KEY environment variable. Run the server via python server.py, fastmcp run server.py, or install it for Claude Desktop with fastmcp install server.py. Two main tools are available: verify_email (single email) and batch_verify_emails (multiple emails with summary statistics).
Key features of AtData SafeToSend MCP Server
- Email verification for individual addresses
- Batch verification with summary statistics
- Comprehensive error handling for various failure scenarios
- Secure API key management through environment variables
- Rate limiting awareness and proper handling
Use cases of AtData SafeToSend MCP Server
- Verifying email addresses before sending marketing campaigns
- Batch-cleaning email lists to improve deliverability
- Integrating email validation into AI assistants via MCP
- Reducing bounce rates by filtering invalid or high‑risk emails
FAQ from AtData SafeToSend MCP Server
What does this server do?
It verifies email addresses using the AtData SafeToSend API, providing status, deliverability, and risk level for each email.
What are the runtime requirements?
Python, plus the dependencies managed by uv or pip. The FastMCP framework is used.
How do I authenticate?
Set the environment variable ATDATA_API_KEY with your AtData API key before running the server.
Can I verify multiple emails at once?
Yes, the batch_verify_emails tool accepts a list of emails and returns per‑result details plus a summary (total, successful, failed, success rate).
What happens if the API rate limit is exceeded?
The server detects HTTP 429 responses and handles them gracefully, returning an appropriate error message.
