Analyzemail MCP Server
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Analyzemail MCP Serververify john@example.com before I send the newsletter"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@analyzemail/mcp
Give your AI agent an email check to run before it sends.
An MCP server for Analyzemail. It lets Claude, Cursor or any MCP client verify an address in real time, clean a whole list file, and download the addresses that are safe to send to, with the reason for every one it removed.
{
"mcpServers": {
"analyzemail": {
"command": "npx",
"args": ["-y", "@analyzemail/mcp"],
"env": { "ANALYZEMAIL_API_KEY": "am_live_..." }
}
}
}Zero dependencies. Node 20 or later. Get an API key at analyzemail.com/account/settings/api. 250 checks a month are free on any account with a confirmed email address.
Tools
Tool | What it does | Cost |
| Checks one address: syntax, the domain's mail servers, whether the mailbox exists, and flags for disposable, role, free, catch-all, protected, parked, spam-trap and litigator addresses. | 1 credit, refunded when the result is |
| Credits the account can spend now, free credits included, and when the free credits reset. | Free |
| Uploads a CSV or TXT file (or reuses a | 1 credit per address |
| Progress of a cleanse; once complete, how many addresses are deliverable, undeliverable, risky and unknown. | Free |
| Writes the finished results to a CSV: the list's original columns plus the verdict for each address. Defaults to the | Free |
The server tells the model how to use them: verify before sending to addresses it did not
collect itself, send only to deliverable or verified, show a person the risky and
unknown ones, and never raise max_credits without the person agreeing.
Related MCP server: Email Verifier MCP Server
Spending limits
An agent should not be able to spend more than you expect.
clean_listuploads the list, reads how many addresses it has, and refuses to start when that is more thanmax_credits(default 1,000) or more than your balance. Nothing is charged; the refusal says the cost and thelist_idto start it with once a person agrees.Accounts that have never bought credits are limited to 2 checks a second and 500 a day. The server waits out short rate limits on its own.
Revoke a key from your account at any time.
A typical run
clean_list { "path": "leads.csv" }
-> list 513, 1,200 addresses, job 9051 running
job_status { "job_id": 9051 }
-> complete, result 878: 1,010 deliverable, 96 undeliverable, 71 risky, 23 unknown
download_results { "result_id": 878, "out_path": "leads.verified.csv" }
-> 1,010 rows writtenA verified download first re-checks verified addresses that were checked more than a week
ago, so verified means deliverable now. While that runs, download_results returns
rechecking with the count; call it again a few minutes later.
What it does not do
It does not send email. It tells your agent which addresses are safe to send to.
It does not make a mailbox exist. A mailbox that exists today can be closed tomorrow, and some mail servers (security gateways, catch-all domains) do not say whether a mailbox exists. Those come back as risky or unknown, not deliverable.
It does not decide about risky addresses. Role accounts, catch-all domains and the like are for a person to judge.
Configuration
Variable | |
| Required. Starts with |
| Optional. Defaults to |
Claude Code: claude mcp add analyzemail --env ANALYZEMAIL_API_KEY=am_live_... -- npx -y @analyzemail/mcp
Related
send-guard, by our partner AskEleven: free local list checks with no account, and mailbox checks through Analyzemail. CLI, MCP server and library.
inbound-guard: the other direction. Scores an incoming email as cold outreach.
API reference and the OpenAPI spec.
License
MIT
Related MCP Connectors
Email verification for AI agents: syntax, MX, disposable, role, score. $0.001 USDC/call via x402.
Verify emails (valid/invalid/risky/unknown): MX, SMTP, catch-all, disposable. Free agent key.
Verify emails (valid/invalid/risky/unknown): MX, SMTP, catch-all, disposable. Free agent key.
Verify emails (valid/invalid/risky/unknown): MX, SMTP, catch-all, disposable. Free agent key.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to verify email addresses through a multi-signal probabilistic pipeline, returning confidence scores and honest statuses (safe/risky/invalid/unknown) with evidence.Apache 2.0
- FlicenseNot gradedqualityBmaintenanceEnables AI agents to verify email addresses in real-time, checking syntax, DNS, MX records, and SMTP handshake, returning verdicts and deliverability scores without sending actual emails.-
- AlicenseAqualityAmaintenanceEnables AI agents to verify email addresses and receive a 'send', 'hold', or 'kill' verdict with the reason, while also managing signup and credits programmatically.52MIT
- AlicenseAqualityBmaintenanceEnables agents to validate email addresses through RFC 5321 syntax checks, MX record lookups, a 182,000-domain disposable blacklist, role-account detection, and an optional SMTP handshake. Runs locally via npx or as a hosted HTTP endpoint, returning explainable per-check scores at zero per-lookup API cost.4MIT