Vercel MCP Server

Vercel's official hosted MCP server (beta): query projects, deployments, build logs, and Web Analytics using natural language from Claude, ChatGPT, and more.

Description

The Vercel MCP Server is the official Model Context Protocol connection to your own Vercel projects, hosted by Vercel itself. Instead of switching between the Vercel dashboard, terminal, and AI assistant, a coding agent connects directly to your Vercel account and can manage projects, analyze deployments and build logs, query Web Analytics data, and search Vercel documentation — all through natural language chat in Claude, ChatGPT, Cursor, or other supported clients.

What the server does

Per the provider, the tool catalog is organized into several areas: documentation tools for targeted searches of Vercel docs by topic, project management tools for listing teams and projects and retrieving project details like framework and domains, deployment tools for listing deployments, retrieving individual deployment details, and analyzing build logs with targeted filtering on error lines, and tools for querying visitors, page views, and custom events from Web Analytics. Each tool is documented with a sample prompt you can send directly to the connected AI assistant, such as "Show me all deployments for my blog project."

Public and authenticated tools

The provider distinguishes between two tool categories: public tools that work without authentication — such as documentation search — and authenticated tools that require a Vercel sign-in to access projects, deployments, and analytics data. This separation allows the server to be used for pure documentation research without linking your own Vercel account.

Supported clients and setup

Per the provider, the server implements the latest MCP Authorization and Streamable HTTP specifications and is reachable at https://mcp.vercel.com. Officially supported clients include Claude Code, Claude.ai and Claude Desktop, ChatGPT, Codex CLI, Cursor, Visual Studio Code with GitHub Copilot, Devin, Raycast, Goose, Windsurf, and Gemini Code Assist and Gemini CLI; per the provider, additional clients will be added over time. For security reasons, Vercel MCP only supports clients that Vercel has previously reviewed and approved. Setup can be automated for many clients with a one-line command via the add-mcp CLI tool, which automatically detects installed AI clients and configures them.

Authentication

Sign-in happens via OAuth through the respective MCP client; a manually created access token is not needed for regular use. After adding the server, the client prompts for authorization and opens a browser window to sign in to your Vercel account.

Status, license, and cost

Per the provider, Vercel MCP is in beta and available on all Vercel plans without separate additional costs; use is subject to Vercel's Public Beta Agreement and AI Product Terms. An existing Vercel account is required.

Security note

Vercel explicitly recommends enabling human confirmation for tool execution and exercising particular caution when using multiple MCP servers simultaneously, to avoid prompt injection attacks.

Who benefits most

The server is especially useful for development teams already hosting applications on Vercel: analyzing build errors directly in the editor, checking deployment status without switching to the dashboard, or evaluating Web Analytics metrics right inside a chat with the AI assistant. For pure documentation research, unauthenticated use of the public tools alone is enough.

Requirements

An existing Vercel account plus an MCP client reviewed and approved by Vercel (including Claude Code, Claude.ai, ChatGPT, Codex CLI, Cursor, VS Code with GitHub Copilot, Gemini CLI). No Vercel account is needed for pure documentation search.

Installation instructions

The Vercel MCP Server is hosted: nothing is installed locally, and the MCP client connects directly to https://mcp.vercel.com.

Step 1: Fastest setup with add-mcp

bash
npx -y add-mcp https://mcp.vercel.com -g

The tool automatically detects installed AI clients and configures Vercel MCP for all of them. Without -y it asks for confirmation before each install; without -g it installs only for the current project instead of globally.

Step 2: Setting it up manually in Claude Code

bash
claude mcp add --transport http vercel https://mcp.vercel.com

Then run /mcp to start the OAuth sign-in.

Step 3: Setting it up in Codex CLI

bash
codex mcp add vercel --url https://mcp.vercel.com

Codex automatically opens the browser for authorization when adding the server.

Setting it up in Cursor, VS Code, and other clients

Cursor and VS Code provide one-click install links in the official documentation. For Claude.ai and Claude Desktop, Vercel can be added as a custom connector under Settings → Connectors with the URL https://mcp.vercel.com (available for Pro, Max, Team, and Enterprise users). For ChatGPT, developer mode must first be enabled under Settings → Connectors → Advanced settings before a new connector with OAuth authentication can be created.

Staying secure

Vercel recommends enabling human confirmation for tool execution rather than letting agents make deployment or project changes unsupervised. When using multiple MCP servers simultaneously, be particularly careful about prompt injection attacks: malicious content from another connected server could try to steer the agent into unwanted actions in Vercel.

Keeping the beta status in mind

Vercel MCP is still in beta and subject to Vercel's Public Beta Agreement. Feature scope and supported clients may therefore change; the official documentation always reflects the current state.

Common connection issues

If add-mcp fails with an error about an unrecognized client, manual setup via the corresponding claude mcp add or codex mcp add command usually works. If no browser window opens during authorization, check whether the client is blocking pop-ups or whether a displayed URL needs to be opened manually.

bash
claude mcp add
codex mcp add

For teams with multiple Vercel projects

Anyone working across multiple Vercel teams or projects should carefully check which team or project is currently connected before making production deployment changes. Since the agent operates with the signed-in account's permissions, a mix-up could otherwise lead to changes on the wrong project.

A quick check of the active context before any major deployment action is far faster than rolling back an accidental change afterward.

Building this check into a standard checklist before agent-driven deployments is worth the small overhead.

Source: vercel.com/docs/agent-resources/vercel-mcp and vercel.com/docs/mcp/vercel-mcp/tools, checked on 2026-09-06.

npx -y add-mcp https://mcp.vercel.com -g

Authentication

OAuth via the respective MCP client; no manually created access token needed for regular use. After adding the server, the client opens a browser window to sign in to your Vercel account.

Required access permissions

Public tools (e.g. documentation search) work without authentication. Authenticated tools for projects, deployments, and analytics act strictly with the permissions of the signed-in Vercel account.

Transmitted or stored data

Requests run through Vercel-hosted infrastructure at mcp.vercel.com. Authenticated responses contain data from your own Vercel account (project configuration, deployment metadata, build logs, Web Analytics metrics) and are returned to the MCP client and its language model.

Security risks

Vercel recommends enabling human confirmation for tool execution and exercising particular caution when using multiple MCP servers simultaneously, to avoid prompt injection attacks. For security reasons, only clients reviewed by Vercel are supported.

License and costs

License
Not recorded yet.
Cost
free

Per the provider, the MCP server is in beta and available on all Vercel plans without separate additional costs. An existing Vercel account is required. Purchasing a Vercel plan happens exclusively on vercel.com, not through Skill Road.

Alternatives

Not recorded yet.

At a glance

Provider
Vercel
Status
Official server
Deployment
Remote
Current version
Not recorded yet.
GitHub stars
18
Last reviewed
06.09.2026

Repository and documentation

Categories

Supported clients