Verifly MCP Server
Enables AI agents to verify email addresses, clean lists, run bulk verification jobs, and manage a Verifly account through natural language, without writing HTTP calls.
README
<p align="center"> <img src="https://raw.githubusercontent.com/james-sib/verifly-mcp-server/master/logo.png" width="120" alt="Verifly logo" /> </p>
Free email tools & API
- Catch-all email verifier — check if a domain is accept-all
- Disposable email checker
- Bulk email verification API — clean CSV lists
- Email verification API for developers — pay-as-you-go, 100 free credits, no monthly fee
Verifly MCP Server
Agent-native email verification over the Model Context Protocol. This server gives any MCP client — Cline, Claude Desktop, Claude Code, Cursor, Windsurf — a clean set of tools to verify email addresses, clean lists, run async bulk jobs, and manage a Verifly account, all without writing a single raw HTTP call.
It is built for autonomous workflows: an agent can self-register for an API key, verify a lead before sending, deduplicate and scrub an import list, kick off a bulk job and poll it to completion, and check its own credits and usage — end to end.
Tools
15 tools cover the full Verifly workflow:
| Tool | What it does |
|---|---|
verify_email |
Verify a single address in real time (verdict, reason, flags, send/reject recommendation, credits) |
verify_batch |
Verify a list synchronously (per-address verdicts; best for up to a few hundred) |
clean_email_list |
Dedupe + drop invalid syntax, disposable, and role addresses from a list |
extract_emails |
Pull every email address out of free-form text (notes, signatures, pasted docs) |
check_domain_health |
MX / SPF / DMARC records + an overall health score for a domain |
get_credits |
Remaining verification credits + recent usage (free, no credits) |
submit_bulk |
Submit an async bulk verification job for large lists (returns a job_id; optional webhook_url) |
get_job_status |
Status + progress of a bulk job |
get_job_results |
Per-address results of a completed bulk job |
list_jobs |
List the account's bulk jobs (filter by status, paginate) |
get_usage |
Detailed usage statistics for a period (day / week / month) |
get_account |
Account profile: email, company, credits, plan, key count (free) |
get_packages |
List the credit packages available to purchase |
buy_credits |
Start a credit-package purchase and return a checkout link |
register_account |
Self-onboard a brand-new account; returns an api_key shown once |
Authentication
Get an API key from https://verifly.email (or have an agent self-register with
register_account). The server reads the key from the VERIFLY_API_KEY
environment variable. Every Verifly request is authenticated with
Authorization: Bearer <key>.
The hosted HTTP transport also accepts a per-request Authorization: Bearer <key>
header (which overrides the env var), so each caller can supply their own key.
Install (local, stdio) — recommended for Cline
npx -y verifly-mcp-server
or install it globally:
npm install -g verifly-mcp-server
verifly-mcp-server
Client config
Add this to your MCP client config (Cline: cline_mcp_settings.json; Claude
Desktop: claude_desktop_config.json):
{
"mcpServers": {
"verifly": {
"command": "npx",
"args": ["-y", "verifly-mcp-server"],
"env": {
"VERIFLY_API_KEY": "vf_your_api_key"
}
}
}
}
That's the whole setup — no build step, no clone. See
llms-install.md for an agent-followable install guide.
Hosted Streamable-HTTP transport
If you'd rather not run anything locally, Verifly hosts the same server as a remote Streamable-HTTP endpoint:
https://verifly.email/mcp
Point any Streamable-HTTP-capable MCP client at that URL and pass your key as an
Authorization: Bearer <key> header. Example raw call:
curl -s -X POST https://verifly.email/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-H "Authorization: Bearer vf_your_api_key" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
"params":{"name":"verify_email","arguments":{"email":"lead@example.com"}}}'
Self-hosting the HTTP transport
PORT=8787 VERIFLY_API_KEY=vf_your_api_key node src/http.js
# or: PORT=8787 VERIFLY_API_KEY=vf_... npm run start:http
| Var | Default | Purpose |
|---|---|---|
PORT |
8787 |
Listen port |
HOST |
0.0.0.0 |
Bind address |
MCP_PATH |
/mcp |
Path the MCP endpoint is served on |
VERIFLY_API_KEY |
— | Fallback key when a request sends no Authorization header |
VERIFLY_BASE_URL |
https://verifly.email |
API base override |
- MCP endpoint:
POST {MCP_PATH}(stateless Streamable HTTP). - Health check:
GET /healthz→{"ok":true}.
Test
VERIFLY_API_KEY=vf_your_api_key npm run test:e2e
Drives the stdio server through the MCP protocol and performs a real
verify_email + get_credits call against the live API.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。