A Purpose-Built MCP Gateway for AI Agents
agentgateway is an MCP gateway that secures and observes every MCP tool call your agents make, with zero-trust security, audit logs, and the policy controls a compliance-ready agent platform needs.
Capability comparison
agentgateway is the only AI gateway with full marks in all 3 categories
Why Teams Choose Us
“We can’t tell which agents are calling which MCP tools, or whether that access was ever approved.”
agentgateway sits in front of every MCP tool call, so you can see, control, and audit tool access the same way you would any other API traffic instead of trusting it blindly.
“Every MCP client and server combination needs its own one-off integration to secure.”
agentgateway is a single MCP gateway layer that sits between agents and tools, so policy and access control apply consistently no matter which MCP servers your teams connect.
“Security and finance won’t sign off on agentic tool access without audit logs and policy controls.”
agentgateway gives you audit logs, access policy, and zero-trust security on every MCP tool call by default, so compliance is built in rather than bolted on after the fact.
Trust & compliance
SOC 2
Certification supports security and procurement review for regulated Kubernetes environments
OSS roots
Build on Envoy, Istio, Cilium, and Gateway API foundations for open source service mesh adoption.
CNCF maintainer status
keeps Solo.io engineers close to the projects shaping cloud native mesh standards.
FIPS builds
provide FIPS 140-2 validated options for service mesh security requirements.
Published SLAs
Define support expectations for enterprise service mesh operations.
Customer references
include production outcomes from teams running large-scale Kubernetes networking.
Frequently Asked Questions
What is an MCP gateway?
What is an MCP gateway?
An MCP gateway sits between AI agents and the MCP (Model Context Protocol) tools and servers they call, securing and observing that traffic the way an API gateway does for regular APIs. agentgateway is Solo's open source, AI-native gateway purpose-built for this, along with LLM access and agent-to-agent (A2A) communication.
Is agentgateway also an LLM gateway?
Is agentgateway also an LLM gateway?
Yes. agentgateway handles LLM traffic natively, including access control, cost visibility, and audit logging for calls to your LLM providers, alongside its support for MCP and A2A protocols.
How does agentgateway secure MCP tool calls specifically?
How does agentgateway secure MCP tool calls specifically?
agentgateway authenticates and authorizes every MCP tool call, logs which agent called which tool with what arguments, and lets you enforce policy on that traffic the same way you would for any other API call, instead of trusting MCP servers blindly.
How is agentgateway different from a cloud provider's AI gateway platform?
How is agentgateway different from a cloud provider's AI gateway platform?
agentgateway is open source and self-hosted, so it isn't tied to one cloud. Unlike a walled-garden ai gateway platform from a single vendor, you can run it anywhere Kubernetes runs without re-architecting when you change providers.
Do I need an enterprise contract to use an open source ai gateway like agentgateway?
Do I need an enterprise contract to use an open source ai gateway like agentgateway?
Most production deployments choose Solo's enterprise support for the SLAs, audit tooling, and production hardening a compliance-ready platform needs. The underlying project is open source and donated to a vendor-neutral foundation under the Linux Foundation, so it also remains free to run on your own if you'd rather self-support.
See the MCP Gateway Built for AI Agents
Request a demo to see how agentgateway secures MCP tool calls in your environment, or compare your options in the Buyer's Guide first.
