WeChat ID
aaronpk_tv
#xaa
-
Figuring out how AI agents get access to enterprise apps gets messy fast.
Static API keys and repeated OAuth flows look fine in a demo, but they completely break down at scale.
Lately, much of my work in the IETF OAuth Working Group has focused on solving this exact bottleneck.
By leveraging the Cross-App Access pattern, built on the Identity Assertion JWT Authorization Grant, we can fundamentally change how agents interact with your stack:
• No more manual OAuth dance: Agents get seamless, scoped access to connected apps entirely behind the scenes.
• Centralized IT control: Enterprise admins get the clear visibility, security boundaries, and policy control they actually need.
I’m joining Jiquan Ngiam, CEO of MintMCP, to discuss how this plays out in practice. JQ runs over 20 agents on MintMCP's platform, so we’ll explore what MCP Enterprise-Managed Authorization looks like when deployed across tools like Salesforce, GitHub, and Confluence.
If you work in identity, security, or are currently figuring out how to safely deploy AI agents in your enterprise, come join the conversation.
Free and live on Zoom, Jul 9 at 9am Pacific: https://luma.com/va1tfrnf -
Enterprise AI just got a lot more secure. Anthropic launched a beta of "Enterprise Managed Auth" in Claude, so you can now connect Claude seamlessly to MCP servers through your enterprise IdP like Okta!
Now employees no longer have to connect MCP servers manually and wait for a series of OAuth and login prompts. Once you log in to Claude from Okta, all the preconfigured MCP servers are already connected! It's not every day you get to improve both usability and security!
This is an application of the Cross App Access pattern, defined in the Identity Assertion JWT Authorization Grant being standardized in the OAuth working group at the IETF.
Seeing adoption from a massive player like Claude is a huge validation of the effort! It's been fantastic to work with the folks at Anthropic over the past year on this Paul Carleton and Den Delimarsky. And of course this wouldn't be possible without the collaboration with my co-authors on the spec Karl McGuinness and Brian Campbell!
https://claude.com/blog/enterprise-managed-auth
https://www.youtube.com/watch?v=5kTDt9ewTwE -
Cross-Domain API Access: Beyond the "Obvious" Shortcuts
Cross-domain access is everywhere in today's software landscape. Whether you look at enterprise SaaS applications, AI agents interacting with user data across multiple platforms, or "integrated experiences" pulling information from a calendar, a chat tool, and a wiki—everything eventually needs to talk across boundaries.continue reading... -
The "Agent Verified" signup flow from WorkOS is exactly what I've been telling the agent platforms they should be doing with Cross App Access! Very cool to see this launch! 👏
https://workos.com/auth-md/docs/flows/verified
"The agent's provider — OpenAI, Anthropic, Cursor, or any trusted agent platform — attests to the user's identity at registration time. Your service verifies the attestation and issues credentials synchronously, no human interaction required."
In Cross App Access terms:
• The "agent platform/provider" is the ID-JAG issuer, because users are already signed in to those platforms when they use agents
• The "service" is the ID-JAG consumer (the Resource AS), and issues an access token if the ID-JAG is trusted and valid
You can test this out in the Cross App Access sandbox today! https://xaa.dev/
