Model Context Protocol
MCP Server
emailvalidation.io runs a hosted Model Context Protocol (MCP) server. AI agents and assistants (Claude, Cursor, VS Code and any other MCP-capable client) can call the API as native tools, no SDK or glue code required.
https://api.emailvalidation.io/mcp
- Hosted, nothing to install
- Streamable HTTP
- Same quota as the API
Setup
Connect your MCP client
All requests, including listing the available tools, need your API key. Send it as the apikey header.
claude mcp add --transport http emailvalidation \ https://api.emailvalidation.io/mcp \ --header "apikey: YOUR-API-KEY"
{
"mcpServers": {
"emailvalidation": {
"url": "https://api.emailvalidation.io/mcp",
"headers": { "apikey": "YOUR-API-KEY" }
}
}
}
Available tools
Each tool maps to one REST endpoint and returns the same JSON and the same machine-readable errors.
getInfo
Validate an email address
GET /v1/infocreateBulkJob
Submit a list for bulk verification
POST /v1/bulklistBulkJobs
List your bulk jobs
GET /v1/bulkgetBulkJob
Get a bulk job’s status
GET /v1/bulk/{job_id}deleteBulkJob
Delete a bulk job
DELETE /v1/bulk/{job_id}getStatus
Account quota status
GET /v1/statusDetails
Every bulk endpoint is a tool except the results download: a completed job’s results can run to 100,000 rows, too large for a tool result, so download them from the job’s links.results over the REST API. createBulkJob takes the list as the emails array; CSV and TXT file uploads are only available over the REST API.
Tool calls are metered like regular API requests and return the same machine-readable errors, so your agent always knows how to proceed. Full details in the MCP documentation.
Email MCP server FAQ
Which AI clients can use the MCP server?
Do MCP tool calls cost extra?
Can an agent validate a whole list?
How does the agent get the results of a bulk job?
Start validating emails for free
100 free validations every month. No credit card required. Upgrade when you need more.