agentgateway

The first complete connectivity solution for Agentic AI.

Agentgateway is an open source data plane optimized for agentic AI connectivity within or across any agent framework or environment. Agentgateway provides drop-in security, observability, and governance for agent-to-agent and agent-to-tool communication and supports leading interoperable protocols, including Agent2Agent (A2A) and Model Context Protocol (MCP).

Intro to Agentgateway Video

Key Features:

• Highly performant: agentgateway is written in rust, and is designed from the ground up to handle any scale you can throw at it.
• Security First: agentgateway includes a robust MCP/A2A focused RBAC system.
• Multi Tenant: agentgateway supports multiple tenants, each with their own set of resources and users.
• Dynamic: agentgateway supports dynamic configuration updates via xDS, without any downtime.
• Run Anywhere: agentgateway can run anywhere with any agent framework, from a single machine to a large scale multi-tenant deployment.
• Legacy API Support: agentgateway can transform legacy APIs into MCP resources. Currently supports OpenAPI. (gRPC coming soon)

Getting Started

To get started with agentgateway, please check out the Getting Started Guide.

Documentation

The agentgateway documentation is available at agentgateway.dev/docs. Agentgateway has a built-in UI for you to explore agentgateway connecting agent-to-agent or agent-to-tool:

Contributing

For instructions on how to contribute to the agentgateway project, see the CONTRIBUTION.md file.

Community Meetings

To join a community meeting, add the agentgateway calendar to your Google account. Then, you can find event details on the calendar.

Recordings of the community meetings will be published on our google drive.

Roadmap

agentgateway is currently in active development. If you want a feature missing, open an issue in our Github repo.

Contributors

Thanks to all contributors who are helping to make agentgateway better.

Star History