Claude 101- Connecting your tools

What are connnectors? 是MCP吧

Key takeaways

  • Connectors transform Claude from an assistant into an informed collaborator
  • Connectors allow Claude to read information and perform actions on your behalf.
  • The Model Context Protocol (MCP) powers connectors.
  • There are two types of connectors: web connectors and desktop extensions.
    • cloud services
    • local on your computer

Finding and connecting tools

  • Anthropic directory at claude.ai/directory
    • Web: Cloud services and applications
      • Find the connector
      • Click Connect:
      • Authenticate
      • Grant permissions:
      • Test the connection: try a simple request
    • Desktop extensions: Local computer
      • Claude Desktop app

Using connectors in your work

  • Project management (Asana, Linear, Jira)

    • Highest-priority tasks, create a new task , Summarize the status
  • Communication (Slack, Gmail)

    • Find the email, Draft a reply, conclusion
  • Documentation (Notion, Google Drive, Confluence)

    • Search, Summarize, QA
  • Business tools (Stripe, PayPal, Salesforce)

    • revenue trends​​​​​​​

Security and permissions

  • Scoped access:individual permissions on and off
  • Claude sees what you see: only your own
  • Revocable at any time: only install connectors from trusted sources.
相关推荐
luckystar513~4 天前
每日AI资讯(2026-09-02)
openai·anthropic·ai动态·ai咨询·每日ai资讯·每日ai动态
云卷云舒___________5 天前
Anthropic发布Fable 5.1与Mythos 5.1登顶,Meta超级应用定名Muse,Spark连发两款端侧智能体 | 9月2日 AI日报
meta·anthropic·ai日报·fable51·mythos51·muse·sparkllm
deepseek2310 天前
Anthropic MHS 拆解:MCP 管软件,MHS 管硬件,AI 第一次安全操作显微镜、机械臂和激光器
硬件·ai agent·anthropic·mcp
Roadinforest15 天前
Claude Code 架构深度解析:从 Agent Loop 到 Tool、MCP 与 Context
ai·架构·llm·agent·anthropic·claudecode
漂流瓶jz21 天前
一文读懂大模型生态:分类/参数/结构/训练/GPU/评测/排行/社区
人工智能·大模型·llm·openai·nvidia·deepseek·anthropic
云卷云舒___________22 天前
Anthropic披露内部模型Model 2、Codex多智能体v2委派多模型、英伟达开源550B推理教师模型 | 8月16日 AI日报
多智能体·codex·英伟达·anthropic·ai日报·model2·推理教师模型
东姬AI24 天前
Anthropic模型36小时突破46年数学进展,同日三家前沿模型上新:当AI抽象智能逼近极限,“表现力“为什么仍是结构性空白
ai·agent·grok·deepseek·anthropic·黎曼猜想·qwen3.8
还是奇怪1 个月前
Agent 安全事件成为本周核心风险信号,从三起越界评测到 Hugging Face 入侵
人工智能·web安全·大语言模型·anthropic
云卷云舒___________1 个月前
视频抽奖ChatGPT Plus | Cursor迭代Composer3、Sonnet5.5或下月发布、腾讯押注WorkBuddy办公AI 8月9日 AI日报
cursor·腾讯·anthropic·vega·ai日报·composer3·sonnet55