> For the complete documentation index, see [llms.txt](https://docs.mcpfunctions.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.mcpfunctions.ai/home/readme.md).

# Introduction

Create MCP tools in seconds. Run them serverless. Connect any AI assistant.

MCP Functions is a platform where you describe what a tool should do, AI generates the code, and the tool is live immediately - no servers, no deploys, no infrastructure. Your tools work with ChatGPT, Claude, Cursor, and every other MCP-compatible client out of the box.

## Start Here

<table data-card-size="large" data-view="cards"><thead><tr><th></th><th></th></tr></thead><tbody><tr><td><a href="/pages/7r7fAiXdGn13Bw3yxDkH"><strong>Set up your account</strong><br>Create organization, workspace, and team access.</a></td><td><a href="/pages/7r7fAiXdGn13Bw3yxDkH"><img src="/files/S8MbfpGFbANpmUeGl79H" alt=""></a></td></tr><tr><td><a href="/pages/dFSio8mwkzk6WZHFJ2Et"><strong>Create your first tool in minutes</strong><br>Build, test, and publish your first working MCP tool.</a></td><td><a href="/pages/dFSio8mwkzk6WZHFJ2Et"><img src="/files/hZCy1UtHQB6mlDMsDX6x" alt=""></a></td></tr><tr><td><a href="/pages/ecumlxfwzhoz6cJMAByW"><strong>Follow step-by-step tutorials</strong><br>Learn by doing with practical, task-based walkthroughs.</a></td><td><a href="/pages/ecumlxfwzhoz6cJMAByW"><img src="/files/H7eQziAl47TQwa2ELuKH" alt=""></a></td></tr><tr><td><a href="/pages/QA0yCEdyOCIo4X6r9tSO"><strong>Connect an AI assistant</strong><br>Connect ChatGPT, Claude, Cursor, or any MCP client.</a></td><td><a href="/pages/QA0yCEdyOCIo4X6r9tSO"><img src="/files/c9SpF5yVg0rDuqcmbbUG" alt=""></a></td></tr></tbody></table>

## How It Works

| Step         | What happens                                               |
| ------------ | ---------------------------------------------------------- |
| **Describe** | Tell the AI what your tool should do in plain language     |
| **Generate** | AI creates the code, parameters, name, and description     |
| **Test**     | Run the tool with sample inputs right in the dashboard     |
| **Publish**  | One click — the tool is live on your MCP server            |
| **Connect**  | Any MCP-compatible assistant discovers and calls your tool |

## Key Capabilities

| Capability             | Details                                                        |
| ---------------------- | -------------------------------------------------------------- |
| **AI tool generation** | Streaming or agent mode, auto-generates name, params, and code |
| **Manual coding**      | Full code editor when you need custom logic                    |
| **Secrets & OAuth**    | Workspace-level encrypted secrets and OAuth2 token injection   |
| **Sandbox execution**  | Isolated VM per call, timeouts, HTTP limits, blocked patterns  |
| **Versioning**         | Publish, roll back, view history                               |
| **Team access**        | Organizations, workspaces, roles (Owner / Admin / Member)      |
| **Dashboard**          | Invocation stats, activity feed, country-level usage           |
| **Full MCP protocol**  | `tools/list`, `tools/call`, `resources`, `prompts`, `logging`  |

## Explore the Docs

| Section                                                                | What you'll find                                         |
| ---------------------------------------------------------------------- | -------------------------------------------------------- |
| [Getting Started](/documentation/getting-started/introduction.md)      | Account setup, first tool, and quick orientation         |
| [Tutorials](/documentation/tutorials.md)                               | Hands-on guides: build, connect, debug                   |
| [Core Concepts](/documentation/core-concepts/platform-architecture.md) | Orgs, workspaces, servers, tools - how they fit together |
| [Creating Tools](/documentation/creating-tools/ai-powered-creation.md) | AI generation, manual creation, and code examples        |
| [Execution](/documentation/execution/execution-flow.md)                | Sandbox, limits, testing, and logs                       |
| [MCP Protocol](/documentation/mcp-protocol/protocol-overview.md)       | Protocol details, connecting clients, tool invocation    |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.mcpfunctions.ai/home/readme.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.