Vercel AI SDK

npm install ai @ai-sdk/mcp

From AI SDK 7 the MCP client lives in @ai-sdk/mcp, and ai no longer exports one. Both packages need Node.js 22 or later.

npx @nextmail/cli login

Choose the browser sign-in. On the approval page, pick the workspace, the access, the addresses the AI SDK may reach and how long the approval lasts. Read only is the default, and it leaves out tools such as sendEmail, replyToEmail and createDraft, so choose Read and send when the agent should draft or reply. The sign-in is saved on this machine and listed in Account → Command line, where you can narrow it or remove it. A sign-in the bridge has not used for 7 days, or one whose approval has ended, has to sign in again.

On a server or over SSH, add --no-browser: open the link on any device, approve, and paste the code it shows back into the terminal. A browser sign-in needs a person once, so a CI job cannot make one.

import { createMCPClient } from '@ai-sdk/mcp'import { StdioMCPTransport } from '@ai-sdk/mcp/mcp-stdio'import { generateText, isStepCount } from 'ai' const client = await createMCPClient({  transport: new StdioMCPTransport({    command: 'npx',    args: ['-y', '@nextmail/cli', 'mcp', 'serve'],  }),}) try {  const { whoAmI, listThreads, getThread } = await client.tools()  const { text } = await generateText({    model: 'openai/gpt-6-astra',    tools: { whoAmI, listThreads, getThread },    stopWhen: isStepCount(10),    prompt: 'Summarise my unread mail.',  })  console.log(text)} finally {  await client.close()}

The framework starts the local bridge, npx -y @nextmail/cli mcp serve, as a stdio MCP server. The bridge sends each call to the MCP server with your saved sign-in and renews the access token when it runs out, so the agent never handles a token. To use another profile of the command line, add --profile <name> to the arguments. The bridge runs on Node.js 20.12 or later.

• —client.tools() returns the tools by name, so taking three out of it is the filter.
• —stopWhen: isStepCount(10) lets the model call tools and read their results for up to ten steps. Without it, generateText stops after one.
• —StdioMCPTransport passes the bridge only a few environment variables, HOME and PATH among them. Put anything else it needs in env, or name a profile with --profile in args.
• —Swap the model for any the AI SDK supports.

The server registers up to 600 tools, and the AI SDK sees the ones both your role and the approval allow. A tool the approval leaves out is never listed, so the model never tries it. Every tool loaded is sent to the model on each turn, so give the agent the ones its job needs. The example starts from whoAmI, listThreads and getThread.

Stdio is for local servers, and an app on a serverless platform cannot start the bridge. It connects to the remote server, https://api.nextmail.sh/mcp, over HTTP instead, with an authProvider: an OAuthClientProvider from @ai-sdk/mcp. The AI SDK runs discovery, registration and the code exchange, and your app supplies storage for the client and its tokens, the redirect to the approval page and the callback that finishes it. The AI SDK repository has a full example under examples/mcp.

const client = await createMCPClient({  transport: {    type: 'http',    url: 'https://api.nextmail.sh/mcp',    authProvider,  },})