|
|
|
We're excited to share some of the updates from our team, but first, we need to acknowledge and apologize for the outage early Thursday morning. We know Clerk is a critical to your application, and even brief downtime is unacceptable.
|
|
|
Read our postmortem → to learn more about the incident, and the steps we're taking to prevent its recurrence.
|
|
|
Next.js MCP Server Integration
|
|
|
|
We’ve introduced an official MCP server implementation built on Next.js, enabling your users to access their Clerk data using MCP clients such as Clause, Cursor, and other AI tools.
|
|
|
|
Rather than requiring separate MCP servers with their own authentication protocols, our approach allows you to add MCP capabilities directly to your existing application through a single API endpoint. This eliminates the overhead of deploying and managing additional services - you can expose your app's functionality to AI tools without architectural complexity.
|
|
|
|
The following snippet demonstrates how an MCP route handler would be configured in your Next.js application that supports users authenticating with Clerk:
|
|
|
|
|
|
This release focuses on Next.js support, with support for more frameworks coming soon. Support for legacy MCP clients is possible by using tools such as mcp-remote.
|
|
|
Learn more about our MCP implementation in the changelog →
|
OAuth provider improvements
|
|
|
|
Over the past few months, we’ve been hard at work extending the capabilities of our OAuth provider. These changes paved the way for Clerk as a supported MCP Authentication provider, as well as make it more adaptable for applications needing an OAuth server.
|
|
|
-
New OAuth consent screen with support for individual scopes
|
-
Support for dynamic client registration, enabling clients to programmatically create an OAuth client
|
-
In-browser token exchange support for public clients like single-page applications and mobile apps
|
|
|
|
|
|
|
Other Features, Fixes & Improvements
|
|
|
|
|
|
|
|
|
|
|
|
|
Resources
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|