Agent access needs a revocation boundary without turning MCP connections into a manual approval queue.
Administrators need clear control over upstream access, while developers need agents to reach authorized tools with little friction.
Who you’d be doing this for
“My security lead asked who can turn this agent off, and I had no clean answer.”
Valentina Nunes · Network Security Administrator
Manages access policies for internal applications and tools used by distributed employees and AI agents.
What is at stake
Administrators lack a clear way to approve and revoke an agent’s upstream tool access. You must weigh fast agent connectivity against an authorization boundary security operators can trust.
Why it isn’t already fixed
Every obvious fix costs something else. That’s the part you’d have to decide.
- agent connectivity vs scoped authority
- fast integration vs revocable access
- shared policy surfaces vs connection-specific controls
- first-customer learning vs release safety
Why Cloudflare
MCP Portals connect AI agents to internal and third-party tools, making scoped access central to secure agent connectivity.
Written with these in mind
Not your kind of problem? 90 more at Cloudflare, or browse every organization.
This is the setup. The work is inside.
Running it puts you in the room: the full situation and its constraints, stakeholders who push back in their own words, and the decisions that are yours to make. What you produce becomes a Day One Plan — work you can show someone instead of describing.