> ## Documentation Index > Fetch the complete documentation index at: https://developers.hubspot.com/docs/llms.txt > Use this file to discover all available pages before exploring further. --- id: 8c917ba0-a6df-444f-b4eb-90348e25a61f --- # Introduction > Use AI tools to build on HubSpot, no coding experience required to get started. Building on HubSpot with AI looks different depending on what you're trying to do. You might want to build a custom interface inside HubSpot, manage account data in bulk, or connect an external AI application to HubSpot's CRM. This section covers HubSpot's AI developer tools, how to choose the right one, and how to get the most out of them. The more you understand HubSpot's building blocks and the more comfortable you are with code, the better and faster your results will be. Coding knowledge lets you review what AI builds, catch mistakes, and iterate with confidence. HubSpot knowledge helps you know what's possible and better frame the problems you want to solve. ## The tools HubSpot provides several tools for AI-assisted development. They're designed for different use cases, and many can be used in combination. ```mermaid theme={null} flowchart TD B{What are you trying to do?} -->|Explore possible solutions| Breeze[Breeze Assistant] B -->|"Build on HubSpot (UI Extensions, sites, agent tools, apps, etc.)"| Dev[Developer MCP server] B -->|Have your AI agent manage records and data in your account| Agent[Agent CLI] B -->|Connect an AI application to HubSpot over OAuth| Remote[remote HubSpot MCP server] ``` ### Breeze Assistant Breeze Assistant is HubSpot's built-in AI chat, available directly in your account. It has context about your account data and knows what features are available in your subscription, so it can answer questions, make recommendations, and take action without any setup. It's the best place to start if you're still figuring out what you want to build, or if you need a quick answer about what HubSpot can already do. [Learn more about Breeze Assistant](/docs/build-with-ai/breeze-assistant) ### Developer MCP server The local Developer MCP server connects your AI coding agent to HubSpot's developer platform, giving it access to HubSpot's documentation, APIs, and project tools. Describe what you want, and your agent writes the code, configures the project, and deploys it to your account. You can build things like a custom display on a contact record, a dashboard page inside HubSpot, or a website module built to your exact specs. [Learn more about the Developer MCP server](/docs/build-with-ai/developer-mcp-server) ### Agent CLI The Agent CLI lets your AI read, update, and manage records in your HubSpot account on your behalf. This is especially useful for bulk changes, cleaning up data, or generating test records when you're building something new. Most builders use it as a supporting tool alongside the Developer MCP server, not on its own. It's a terminal-based tool, but your agent handles all the terminal interaction, so you don't need to use the terminal yourself. [Learn more about the Agent CLI](/docs/build-with-ai/agent-cli) ### Remote HubSpot MCP server The remote HubSpot MCP server is a publicly hosted MCP endpoint that external AI tools and applications connect to. It handles OAuth authentication and gives the connecting tool access to perform CRM operations in the authorizing user's account. This is useful if you're using an AI tool that doesn't have a HubSpot connector, or if you're building an application that needs to read or write HubSpot data. If you're already using Claude, ChatGPT, Microsoft Copilot, or Gemini, HubSpot offers built-in connectors for those tools, so no custom setup is needed. [Learn more about the HubSpot MCP server](/docs/build-with-ai/remote-mcp-server) ## Next steps Foundational concepts and tips for getting better results from AI tools. Follow the quickstart guide to set up your agent tooling and start building on HubSpot. Ready-to-use prompts for common HubSpot builds. Just paste and go.