mcp-hubspot

概述

mcp-hubspot 服务器通过模型上下文协议 (MCP) 实现使 AI 助手能够直接与 HubSpot CRM 数据交互。它提供管理联系人、公司、检索活动数据以及在之前检索的数据上执行语义搜索的工具。该实现使用 FAISS 的内置向量存储和嵌入缓存来提高性能并克服 HubSpot API 的限制。

试试问 AI

装完之后，这里有 5 个你可以让 AI 做的事：

• hubspot_create_contact

  Create contacts with duplicate prevention

  试试:Create a new contact in HubSpot试试:Add this person to my contacts list
• hubspot_create_company

  Create companies with duplicate prevention

  试试:Create a new company in HubSpot试试:Add this organization to my CRM
• hubspot_get_company_activity

  Retrieve activity for specific companies

  试试:Show recent activity for Acme Corp试试:What's happening with this company?
• hubspot_get_active_companies

  Retrieve most recently active companies

  试试:Which companies have been most active recently?试试:Show me top engaged companies
• hubspot_get_active_contacts

  Retrieve most recently active contacts

  试试:Which contacts have been most active recently?试试:Show me my top engaged contacts
• hubspot_get_recent_conversations

  Retrieve recent conversation threads with messages

  试试:Show me recent conversations试试:What did we talk about with this contact?
• hubspot_search_data

  Semantic search across previously retrieved HubSpot data

  试试:Find deals related to AI试试:Search for customers interested in SaaS

可对比工具

安装

npx -y @smithery/cli@latest install mcp-hubspot --client claude

docker run -e HUBSPOT_ACCESS_TOKEN=your_token buryhuang/mcp-hubspot:latest

{
  "mcpServers": {
    "hubspot": {
      "command": "docker",
      "args": [
        "run", "-i", "--rm",
        "-e", "HUBSPOT_ACCESS_TOKEN=your_token",
        "-v", "/path/to/storage:/storage",
        "buryhuang/mcp-hubspot:latest"
      ]
    }
  }
}

FAQ

这个 MCP 服务器中的向量存储有什么用途？

使用 FAISS 的向量存储能够对先前检索的 HubSpot 数据进行语义搜索，使 AI 助手能够基于含义而非关键字查找相关信息。

这个服务器如何处理 HubSpot API 限制？

该实现包括嵌入缓存和线程级索引，以克服 API 速率限制并提高响应时间，同时在会话之间保持数据持久性。

mcp-hubspot 对比