Email Spam Tester

io.github.serg-tanichev由 GitHub 帳號 serg-tanichev 發布v1.0.0更新於 Nov 2, 2026

Test an email before it goes out: a disposable address, 41 checks with RFC citations, a fix plan.

已驗證Streamable HTTP可網頁執行溝通與協作商業與電商

概覽

AI 產生的概覽

遠端 MCP 伺服器，用來在寄出前測試郵件的送達率：提供一次性測試地址，並回傳含修正方案的評分報告。

可以做什麼
  • 預留一次性測試地址，可選擇同時檢查各信箱服務商的投遞位置，並取得報告連結
  • 阻塞等待分析完成，回傳評分、未通過與警告的檢查項目以及修正方案
  • 依 slug 直接讀取既有報告，不必等待
  • 取得郵件實際送達的原始內容，包含標頭、兩個內文部分與原始原始碼
適用情境
適合在寄送行銷或交易性郵件前檢查其垃圾郵件與送達表現，並讓助理負責預留地址、等待結果並說明修正建議。對從個人筆電寄出的郵件幫助有限，因為許多檢查讀取的是寄送平台加入的標頭。
執行需求
使用遠端 streamable HTTP 端點，本機無需安裝；不需要金鑰或帳號。不支援 streamable HTTP 的用戶端可用 mcp-remote 橋接。需由本人從實際寄送郵件的平台寄出真實郵件到預留地址。
安裝前請注意
報告 slug 就是唯一憑證：持有者即可讀取報告，包括主旨、寄件者、退信地址與完整原始碼，未經同意不要測試他人的郵件。郵件原始碼在送達後保留一小時即刪除。投遞位置檢查使用部分公開測試信箱，請勿用於機密郵件。skip 狀態不等於通過，complete 為 false 時評分偏樂觀。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Email Spam Tester，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.serg-tanichev%2Femail-spam-tester/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "email-spam-tester": {
      "type": "http",
      "url": "https://email-spam-tester.com/mcp"
    }
  }
}

README

Email Spam Tester MCP server

Email Spam Tester is a free, independent email deliverability testing tool built by Serhii Tanichev. This repository is the client side of its MCP server: how to connect an agent to it, what the tools do, and how to read what they return. There is nothing to install on a server; the MCP endpoint is hosted with the tool.

The service exposes an MCP server at https://email-spam-tester.com/mcp over streamable HTTP. No key, no account. The one-hour test works in every client:

ToolWhat it does
get_test_address(lang, include_placement=false)Reserves a disposable address and returns it with the slug and the report URL. lang decides the language of the fix plan a person sees on the report page. With include_placement the same send also shows which folder the message lands in at different mailbox providers: you get one list of recipients and a marker for the subject. Some of those inboxes are public test inboxes, so use it for mail that is not confidential.
wait_for_report(slug, timeout_seconds=150)Blocks until the analysis is done and returns a summary: scores, the failing and warning checks with their evidence, the fix plan. Nothing to poll and no sleep loop to get wrong.
get_report(slug)Reads an existing report without waiting.
get_message_source(slug)The message as delivered: headers in wire order, both bodies, the raw source. It is kept for an hour after it arrives, and in that hour anyone with the slug can read it; after that only the report remains.

get_report also returns provider placement so far, when the test asked for it.

In ChatGPT there is more, because ChatGPT tells the server which of its users is asking. Permanent mailboxes collect a series of letters (a welcome sequence, a newsletter) and check every one: open_series_mailbox(domain), list_series_mailboxes(), get_series_letters(mailbox_id), write_fix_plan(slug) and close_series_mailbox(mailbox_id). Other clients, Claude among them, are not offered these tools: a permanent mailbox belongs to someone, and they don't say who is asking. Every tool also returns structured content for an MCP Apps card (ui://widget/email-spam-tester-v3.html): the address with copy buttons that turns into the report when the mail lands. Clients that draw no card read the same result as text.

The agent's job in between is to send a real message to the address, from the platform that will send the campaign. Half the checks read headers the sending platform adds, so a copy sent from a laptop says little about it.

Claude plugin

This repository is also a Claude plugin and its own marketplace. In Claude Code:

bash
/plugin marketplace add serg-tanichev/email-spam-tester-mcp/plugin install email-spam-tester@email-spam-tester

The plugin is two things and nothing else. .mcp.json connects Claude to the remote server at https://email-spam-tester.com/mcp, and skills/email-spam-tester/SKILL.md tells Claude how to run a test: reserve an address, have the real message sent to it, wait for the report, fix, test again. There is no local code, no hook and no script. What leaves your machine is what the tools send to that server (a report slug, a language code, whether you asked for provider placement), and the email you choose to send to the test address. What the service keeps, and for how long, is in the privacy policy.

Claude Desktop

Add the server to claude_desktop_config.json (see claude-desktop.json for the whole file):

json
{  "mcpServers": {    "email-spam-tester": {      "url": "https://email-spam-tester.com/mcp"    }  }}

Clients that do not speak streamable HTTP directly can bridge it with mcp-remote:

json
{  "mcpServers": {    "email-spam-tester": {      "command": "npx",      "args": ["-y", "mcp-remote", "https://email-spam-tester.com/mcp"]    }  }}

ChatGPT

The ChatGPT plugin package (manifest, icons, review test cases) is in serg-tanichev/email-spam-tester-chatgpt-plugin. Until it is in the plugin directory you can add the server yourself: Plugins → Add → Create MCP App, server URL https://email-spam-tester.com/mcp, no authentication.

Claude Code

bash
claude mcp add --transport http email-spam-tester https://email-spam-tester.com/mcp

Reading what comes back

  • The per-check lines the tools return are in English whatever lang was set to; lang decides what the person opening the report URL sees.
  • A check status of skip is not a pass: the check did not apply. error means it could not run and is left out of the score.
  • complete: false means something could not be checked. Treat the score as optimistic until it is true.
  • The slug is the capability: whoever holds it can read the report, including the subject, the sender, the bounce address and the full source. Do not test somebody else's mail without asking.

The same instructions, written for an agent to read before it starts, are in SKILL.md, and in the project repository serg-tanichev/email-spam-tester.

Where it is listed

Files

  • claude-desktop.json: a complete Claude Desktop configuration with the server added.
  • server.json: the manifest in the MCP registry format, for registries and directories that read one.

License

MIT, see LICENSE. The license covers this repository, not the service.

來源：README.md，提交 4f2d7c2

工具

4

get_test_address

Get a test address

Reserve a single-use address to send a draft email to, for a full check. Returns the address and the URL its report will appear at. The address accepts exactly one message and expires after an hour. The email has to be sent by the person, from the platform or mailbox they really send from: half the checks read what that platform adds. The report covers 42 checks (SPF, DKIM, DMARC, reverse DNS, TLS, blocklists, content, links, the Gmail and Yahoo bulk-sender rules) and a plan of what to fix. lang is decided now because the message arrives minutes later over SMTP, which carries no hint of who is waiting for it. What wait_for_report returns quotes check lines in English whatever it is set to; the person following report_url sees their own language.

輸入參數langinclude_placement

wait_for_report

Wait for the report

Wait until the message arrives and has been analysed, then summarise it. Call it after the email has been sent to the address from get_test_address. Blocks rather than returning "not ready yet": an agent told to poll writes a sleep loop and usually gives up before the analysis finishes. Provider placement keeps filling in for up to 30 minutes after the report; get_report reads the latest.

輸入參數slug必填timeout_seconds

get_report

Read a report

Read a report by its slug without waiting, with provider placement so far. The report stays at its link for good and anyone with the link can read it: the checks, the score, the fix plan, and the subject and sender of the email.

輸入參數slug必填

get_message_source

Read the received message

The message exactly as it was received: headers, both bodies, source. Useful when a finding points at a header and the agent needs to see what it actually sent rather than what it believes it sent. The message is kept for an hour after it arrives and then deleted; within that hour anyone with the slug can read it. After that only the report remains.

輸入參數slug必填

版本歷史

1
  1. v1.0.0最新Sep 17, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

    請在儲存庫或註冊中心條目中宣告授權條款。