> For the complete documentation index, see [llms.txt](https://docs.patagon.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.patagon.ai/en/getting-started/connect-patagonai-mcp.md). # Connect Patagon AI to your AI assistant Patagon AI MCP lets an AI assistant securely look up the Patagon AI information you already have access to. You can ask about leads, conversations, agents, performance, and reports in everyday language. ## Before you start You need: * An active Patagon AI account. * Access to at least one organization in Patagon AI. * The Patagon AI MCP address: ``` https://api.patagon.ai/mcp ``` When asked to sign in, use your normal Patagon AI login. You do not need an API key or access token. > Patagon AI MCP follows the same access rules as the Patagon AI app. You will only see the organizations and information allowed by your role. If you are limited to assigned leads or conversations in the app, the same limit applies here. Protected contact details remain masked until they are available to you in Patagon AI. ## ChatGPT ChatGPT workspace administrators first need to make Patagon AI available as a custom app. If you do not see the options below, ask your ChatGPT administrator to complete these steps. ### Add Patagon AI to the workspace 1. Open **ChatGPT** in your browser. 2. Open **Workspace settings**. 3. Go to **Apps**, then select **Create**. 4. If prompted, enable **Developer mode**. 5. Name the app `Patagon AI`. 6. Enter `https://api.patagon.ai/mcp` as the MCP server address. 7. Choose **OAuth** authentication. 8. Select **Scan tools** and complete the Patagon AI sign-in window. 9. When the scan finishes, select **Create**. 10. Publish or enable the app for the people who should use it. ### Connect as a workspace member 1. Open **Settings → Apps**. 2. Find **Patagon AI** and select **Connect**. 3. Sign in to Patagon AI and approve the connection. 4. Start a new chat. Select **+ → More**, then enable **Patagon AI**. You can also mention `@Patagon AI` in your prompt if it appears in the app list. ## Codex app 1. Open the **Codex app**. 2. Open **Settings → MCP servers**. 3. Select **Add server**. 4. Enter `Patagon AI` as the name. 5. Choose **Streamable HTTP**. 6. Enter `https://api.patagon.ai/mcp` as the server address and save it. 7. Select **Restart** when prompted. 8. Return to **Settings → MCP servers** and select **Authenticate** next to Patagon AI. 9. Sign in to Patagon AI and approve the connection. You can type `/mcp` in a Codex conversation to confirm that Patagon AI is connected. ### Codex terminal alternative If you use Codex from a terminal, copy and run these commands: ```bash codex mcp add patagonai --url https://api.patagon.ai/mcp codex mcp login patagonai ``` Complete the Patagon AI sign-in in your browser, then start a new Codex conversation. ## Claude Code Claude Code runs in a terminal, but setup only requires one command. 1. Copy and run: ```bash claude mcp add --transport http --scope user patagonai https://api.patagon.ai/mcp ``` 2. Open Claude Code by running `claude`. 3. Type `/mcp`. 4. Select **Patagon AI**, then choose the option to authenticate. 5. Sign in to Patagon AI in the browser window and approve the connection. After setup, Patagon AI is available to your Claude Code conversations on that computer. ## Claude Cowork ### Personal Claude plans 1. Open Claude and go to **Customize → Connectors**. 2. Select **+**, then **Add custom connector**. 3. Name it `Patagon AI`. 4. Enter `https://api.patagon.ai/mcp` as the remote MCP server address. 5. Select **Add**, then sign in to Patagon AI and approve the connection. 6. Open Cowork. Use the **+** button, select **Connectors**, and enable **Patagon AI** for your conversation. ### Claude Team or Enterprise plans An Owner first needs to add Patagon AI for the organization: 1. Go to **Organization settings → Connectors**. 2. Select **Add**. 3. Hover over **Custom**, then select **Web**. 4. Enter `https://api.patagon.ai/mcp` and select **Add**. Each member can then go to **Customize → Connectors**, find **Patagon AI**, select **Connect**, and complete the Patagon AI sign-in. In Cowork, enable it for a conversation from **+ → Connectors**. ## Simple things to try Start with one of these prompts: * `Which Patagon AI organizations can I access?` * `Show me the leads assigned to me that need attention.` * `Summarize the latest conversation with this lead.` * `How are my agents performing? Explain it simply.` * `Give me a short sales manager brief for this week.` * `Create a simple marketing performance summary.` If you belong to more than one organization, include the organization name in your request. You can also ask the assistant to show your organizations first. ## If the connection does not work * **Patagon AI does not appear:** Confirm that the connector is enabled for the current chat or Cowork task. * **You cannot add the connector:** Your ChatGPT or Claude workspace administrator may need to enable custom apps or connectors. * **The sign-in window did not open:** Return to the MCP or connector settings and select **Authenticate** or **Connect** again. * **The browser shows a callback error:** Return to the AI assistant. The sign-in may already be complete; otherwise, retry authentication. * **Information is missing:** Check whether you can see the same organization, lead, or conversation in the Patagon AI app. MCP uses the same permissions. * **Contact details are hidden:** This is expected when Patagon AI's contact-protection rules apply. To disconnect, remove Patagon AI from the assistant's MCP or connector settings. You can reconnect at any time with the same address. ## Official setup references * [Apps in ChatGPT](https://help.openai.com/en/articles/11487775-connectors-in) * [Developer mode and MCP apps in ChatGPT](https://help.openai.com/en/articles/12584461-developer-mode-apps-and-full-mcp-connectors-in-chatgpt-beta) * [MCP in Codex](https://learn.chatgpt.com/docs/extend/mcp) * [MCP in Claude Code](https://code.claude.com/docs/en/mcp) * [Custom connectors in Claude and Cowork](https://support.claude.com/en/articles/11175166-get-started-with-custom-connectors-using-remote-mcp) --- # 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.patagon.ai/en/getting-started/connect-patagonai-mcp.md?ask=&goal= ``` `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.