> For the complete documentation index, see [llms.txt](https://docs.stackai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.stackai.com/stackai-mcp/connect-and-choose-your-organization.md).

# Connect and choose your organization

Connect StackAI to your assistant so you can find projects, inspect workflows, and work with your organization's resources from a conversation. Start with a read-only task to confirm the connection and organization.

## Before you connect

You need a StackAI account with access to the intended organization. Your assistant must support a remote MCP server over HTTP and OAuth. Client plan and workspace policies can restrict custom connections.

Use this server URL:

```
https://mcp.stack.ai/mcp
```

Sign in through the browser authorization flow. Do not paste a password or API key into a chat prompt. These instructions use the MCP server's OAuth connection; REST API authentication is a separate setup.

## Claude

For an individual account, open **Customize → Connectors**, choose **+ Add → Add custom connector**, and name the connection **StackAI**. Enter the URL above and continue through the detected authentication settings. Use OAuth sign-in and complete the StackAI authorization.

For Team or Enterprise, an authorized organization owner must first add the remote connector in **Organization settings → Connectors**. Members then find it in **Customize → Connectors** and select **Connect** to sign in individually.

In your conversation, open **+ → Connectors** and enable StackAI. Availability and owner requirements can vary by account; see [Claude's current custom-connector instructions](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp).

## ChatGPT

Use ChatGPT on the web. Full MCP write actions require an eligible workspace and administrator-enabled access. If you cannot create an app, ask your workspace administrator to make the StackAI connection available.

With developer-mode access, open **Settings → Apps → Create**; administrators can also use **Workspace settings → Apps → Create**. Enter the StackAI MCP URL and select OAuth authentication. Choose **Scan Tools**, finish the StackAI sign-in prompt, then choose **Create**.

Open a chat and select the new app from the tools menu, or mention the app in your request. For workspace-wide availability, an administrator must publish the app and grant access. Tool availability can be restricted by workspace action controls. See [ChatGPT's current MCP app instructions and plan requirements](https://help.openai.com/en/articles/12584461-developer-mode-and-mcp-apps-in-chatgpt).

## Claude Code

Add the remote server:

```bash
claude mcp add --transport http stack-ai-prod https://mcp.stack.ai/mcp
```

Open Claude Code and use `/mcp` to authenticate the connection. Follow the browser sign-in prompt. See [Claude Code's MCP connection instructions](https://code.claude.com/docs/en/mcp).

## Confirm your first result

Ask your connected assistant:

> Confirm my StackAI connection capabilities and active organization. List the projects I can access, with their names, links, and publication status. Do not edit or run any project.

The assistant should call `server_info`, `whoami`, and `projects_list`. Check that the organization name is correct and that project links open the resources you expect. An empty list can be valid if you have no accessible projects.

If the organization is wrong, ask the assistant to select an organization returned by `whoami` using `switch_org`, then call `whoami` again. Keep the discovered organization and resource IDs together.

## If something is missing

* **No StackAI tools:** confirm that the connection is enabled in the conversation and that OAuth sign-in completed.
* **Read tools work but edit/run tools are missing:** inspect `server_info.read_only_mode` and `capabilities.destructive_tools`. A read-only server connection cannot be changed by asking the assistant to write. Ask the connection administrator or StackAI support about write access. If the server supports writes, check the client's action controls and your StackAI permissions.
* **A project is missing or access is denied:** check the active organization and permissions returned by `whoami`. Ask an organization administrator for access to the specific resource.
* **Previously available tools changed:** refresh the client's tool definitions using its app-management controls. In a managed workspace, an administrator may need to approve the update.

## Next steps

Use [Starter prompts](/stackai-mcp/starter-prompts.md) for tasks you can describe in a conversation. Use [Workflow recipes](/stackai-mcp/workflow-recipes.md) when you need to control the sequence of tool calls.

[Best practices and troubleshooting](/stackai-mcp/best-practices-and-troubleshooting.md)


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.stackai.com/stackai-mcp/connect-and-choose-your-organization.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.