mcp-server-apollo

mcp-server-apollo

MCP server for Apollo.io B2B data discovery and enrichment, enabling company search, people search, contact enrichment, and organization firmographics through Apollo's REST API.

Category
访问服务器

README

mcp-server-apollo

MCP (Model Context Protocol) server for the Apollo.io B2B discovery and enrichment API: company search, people search, contact enrichment (emails/phones), organization firmographics, and API usage stats.

The server wraps Apollo's REST API with one account-level API key kept server-side. It is read-only by design: no sequences, no emailer, no CRM writes, no account management.

Tools

Tool Risk Apollo endpoint
apollo_search_capabilities read — (local discovery)
apollo_companies_search read + 1 credit/page POST /mixed_companies/search
apollo_people_search read (free, master key) POST /mixed_people/api_search
apollo_person_enrich read + credits POST /people/match
apollo_people_bulk_enrich read + credits POST /people/bulk_match (chunked ×10)
apollo_org_enrich read + 1 credit GET /organizations/enrich
apollo_webhook_result read (free) GET /webhook_result/{request_id}
apollo_credit_usage read (free, master key) POST /usage_stats/api_usage_stats

Notes that save you a support ticket:

  • People search returns no contact data. Apollo's current mixed_people/api_search endpoint deliberately obfuscates last names and omits emails/phones. Feed the returned person id into apollo_person_enrich to get contact data (that is what costs credits).
  • Phone reveals are asynchronous. reveal_phone_number=true requires a public HTTPS webhookUrl; Apollo POSTs the numbers there minutes later. The synchronous response carries a request_id — poll apollo_webhook_result with it if the callback is missed (results kept 30 days).
  • Master API key. apollo_people_search and apollo_credit_usage require a master Apollo API key (Apollo returns 403 API_INACCESSIBLE otherwise).
  • No credit-balance API. Apollo only exposes per-endpoint rate-limit usage; the remaining credit balance lives in the Apollo UI (Settings → Billing).
  • Field projection. Every data tool accepts fields[] dot-paths to shrink responses; company search applies a slim default projection (pass ["*"] for full records).

Usage

npm install
npm run build

# stdio transport (Claude Desktop, local MCP clients)
APOLLO_API_KEY=... node dist/transports/stdio.js

# streamable HTTP transport on :3000/mcp
APOLLO_API_KEY=... MCP_HTTP_TOKEN=... node dist/transports/http.js

Configuration (see .env.example): APOLLO_API_KEY (required), APOLLO_TIMEOUT_MS, APOLLO_AUDIT_LOG (JSONL audit trail; hashed targets, never raw arguments or secrets), and for HTTP: MCP_HTTP_HOST, PORT, MCP_HTTP_TOKEN (static bearer), MCP_ALLOWED_ORIGINS, MCP_MAX_BODY_BYTES.

There is also an embeddable gateway export (mcp-server-apollo/gateway) that exposes the same tools as plain functions with JSON-schema definitions.

Security

  • The API key is read only from the server environment, sent only in the x-api-key header over HTTPS, and redacted from error messages and logs.
  • Tool calls pass a policy allowlist; anything unlisted is denied.
  • All tools are annotated read-only/non-destructive; credit-consuming tools state their cost in the description so agents confirm spend with the user.
  • HTTP transport binds to loopback by default, validates Origin, caps body size, and supports a static bearer token for upstream auth proxies.
  • Rate limits: HTTP 429 is retried with jittered backoff honouring retry-after; bulk enrichment runs at low bounded concurrency because Apollo limits bulk_match hard (documented 20/min).

Development

npm run typecheck
npm test
npm run smoke:live   # needs a real APOLLO_API_KEY; uses ~1 credit

License

Apache-2.0

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选