# Vercel Connect now supports TanStack AI

Source: [Vercel](https://vercel.com/changelog/vercel-connect-tanstack-ai)  
Feed7 permalink: https://feed7.dev/p/vercel-connect-tanstack-ai-1owhrob  
Published: 2026-09-24T00:00:00.000Z  
Trust: Official Source (official_source)

## Why Included

TanStack AI agents can reach OAuth-protected MCP servers through Vercel Connect, with per-user token refresh and consent handled before model execution.

## Source Summary

The **@vercel/connect/tanstack-ai** entry point adds **connectMCPTransport**, which attaches Vercel Connect authentication to a TanStack MCP transport and refreshes the token before each request.

## Practical Implication

Catch **getConsentChallenge** around createMCPClient and redirect to the supplied consent URL. Handling this before the model runs prevents authorization failures from becoming tool-error text in the conversation.

## Agent-Ready Context

The **@vercel/connect/tanstack-ai** entry point adds **connectMCPTransport**, which attaches Vercel Connect authentication to a TanStack MCP transport and refreshes the token before each request.

Catch **getConsentChallenge** around createMCPClient and redirect to the supplied consent URL. Handling this before the model runs prevents authorization failures from becoming tool-error text in the conversation.

The material demonstrates an HTTP Linear MCP client with per-user identity, but does not describe support for other transport types, deployment environments, or failure recovery beyond consent.

## Connected Context

Feed7 judgment across 875 accumulated Signals:

This makes user-scoped MCP authorization an explicit pre-agent control flow in TanStack AI: attach refreshed credentials to each request and resolve consent before tool execution. It narrows the integration guidance to HTTP MCP and one demonstrated service, leaving transport portability, retry behavior, and recovery from non-consent authentication failures open.

- [Expanding Managed Agents in Gemini API: background tasks, remote MCP and more](https://feed7.dev/p/expanding-managed-agents-gemini-api-1idue29) — Gemini’s remote MCP and credential rotation show the same requirement at a managed-runtime layer, while Vercel Connect specifies the client-side consent interception needed before model execution.
- [Chat SDK now supports reactions and ephemeral messages on Teams](https://feed7.dev/p/chat-sdk-reactions-and-ephemeral-messages-on-teams-1e9k504) — The Teams adapter likewise treats authentication and private permission interactions as integration-specific concerns, reinforcing that a shared SDK abstraction does not eliminate channel- or transport-specific authorization flows.
- [ACP: The Universal Remote Control for AI Agents — Alex Hancock, Block](https://feed7.dev/p/acp-the-universal-remote-control-for-ai-agents-alex-hancock-block-1gvsj4j) — ACP separates client-to-agent sessions from MCP’s agent-to-tool boundary; Vercel Connect clarifies that authentication and consent must still be handled explicitly at the MCP boundary even when the surrounding agent client is portable.

## Context Map

- Layer: tools
- Domains: coding
- Topics: mcp, agent-sdks, tool-use

## Uncertainty

- The material demonstrates an HTTP Linear MCP client with per-user identity, but does not describe support for other transport types, deployment environments, or failure recovery beyond consent.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.