agentgateway | Agent Connectivity SolvedSkip to content

Docs<br>Standalone<br>Kubernetes<br>Tutorials<br>Standalone<br>Kubernetes<br>Models<br>Blog<br>Enterprise<br>CommunityGet Started<br>GitHub

One high-performance gateway for service, LLM, and MCP traffic.<br>An open source HTTP and gRPC gateway that handles traditional application traffic and AI-native protocols in one data plane. Route, secure, observe, and govern services, LLM provider traffic, MCP tools, and agent-to-agent communication without stitching together separate gateways.<br>Part of

Get Started →

See it in action

View on GitHub

Discord

LLM Gateway<br>Route to OpenAI, Claude, Gemini, or any self-hosted model with credentialing and failover.

MCP / A2A<br>Native protocol support for Model Context Protocol and agent-to-agent traffic.

APIs & Services<br>HTTP, gRPC, and TCP traffic — observable by default with mTLS and OIDC built in.

SOURCES · CLIENTSAGENTGATEWAY · CONTROL PLANEDESTINATIONS · MODELS & AGENTSCLI / IDEClaude, Codex, OpenCodeAI agentsLangSmith, CrewAI, ADKApps & clientsAI agentsOpenAIAnthropicMCP / A2AAPIs / ServicesTraffic ManagementLLM GatewayInference RoutingMCP / A2A gatewayREQUEST · INBOUNDROUTE · SECURE · OBSERVE · GOVERNRESPONSE · OUTBOUND

Contributing Companies<br>Linux Foundation<br>Solo.io<br>Microsoft<br>Apple<br>Alibaba<br>Adobe<br>AWS<br>Cisco<br>Salesforce<br>Huawei<br>Amdocs<br>Linux Foundation<br>Solo.io<br>Microsoft<br>Apple<br>Alibaba<br>Adobe<br>AWS<br>Cisco<br>Salesforce<br>Huawei<br>Amdocs

Everything you need for agent traffic — in one binary.<br>One control plane for service traffic, LLM routing, MCP, A2A, and inference — with policy and observability that the platform team actually trusts.

Service Gateway<br>HTTP, gRPC, TCP — observable by default.

Run any north–south or east–west traffic through a single binary. Zero-config TLS, OIDC, mTLS-rotation, native Envoy compatibility for those who want it.<br>HTTP/2gRPCmTLSTLS-1.3<br>Standalone docs &rarr;Kubernetes docs &rarr;

LLM Gateway<br>Route, fail over, and budget across 12+ model backends.

Drop-in OpenAI-compatible API in front of OpenAI, Anthropic, Bedrock, Gemini, Vertex, Cohere, OSS Llama runs — with per-team token budgets, semantic caching and prompt redaction.<br>OpenAIAnthropicBedrockGeminiVertex<br>Standalone docs &rarr;Kubernetes docs &rarr;

Inference Routing<br>Smart routing across self-hosted GPU pools.

Latency-aware, cost-aware, model-aware routing for self-hosted inference. Plug-in vLLM, TGI, Triton; agentgateway picks the warmest replica.<br>vLLMTGITriton<br>Standalone docs &rarr;Kubernetes docs &rarr;

MCP Gateway<br>A control plane for the Model Context Protocol.

Discover, sign, scope and observe every tool call your agents make. Treat MCP servers like microservices — with versioning, RBAC, and audit trails out of the box.<br>DiscoveryRBACAudit<br>Standalone docs &rarr;Kubernetes docs &rarr;

A2A Gateway<br>Bridge agents written in any framework.

agentgateway speaks the Agent-to-Agent protocol natively. Route invocations between LangChain, CrewAI, ADK and your own runtime — with identity, tracing, and replay.<br>LangChainCrewAIADKIdentity<br>Standalone docs &rarr;Kubernetes docs &rarr;

Security & Observability<br>Per-call traces, logs, and policy decisions.

OpenTelemetry by default. Per-tool, per-agent, per-tenant counters. OPA-evaluated allow/deny on every hop, with tamper-evident audit logs.<br>OTELOPAAuditPII-shield<br>Standalone docs &rarr;Kubernetes docs &rarr;

Getting started

View all docs →<br>Install script<br>config.yaml<br>Kubernetes<br>Docker<br>Copy<br># Install the agentgateway binary<br>$ curl -sL https://agentgateway.dev/install | bash

# Run with a config file<br>$ agentgateway -f config.yaml

# INFO agentgateway: Listening on 0.0.0.0:3000<br># INFO agentgateway: Admin UI at http://localhost:15000/ui/

# yaml-language-server: $schema=https://agentgateway.dev/schema/config<br>binds:<br>- port: 3000<br>listeners:<br>- routes:<br>- policies:<br>cors:<br>allowOrigins: ["*"]<br>allowHeaders: [content-type, cache-control]<br>a2a: {}<br>backends:<br>- host: localhost:9999

# Install Gateway API CRDs<br>$ kubectl apply --server-side --force-conflicts \<br>-f https://github.com/kubernetes-sigs/gateway-api/releases/download/v1.5.0/standard-install.yaml

# Install agentgateway CRDs via Helm<br>$ helm upgrade -i agentgateway-crds \<br>oci://cr.agentgateway.dev/charts/agentgateway-crds \<br>--create-namespace --namespace agentgateway-system \<br>--version v1.1.0 \<br>--set controller.image.pullPolicy=Always

# Install agentgateway via Helm<br>$ helm upgrade -i agentgateway \<br>oci://cr.agentgateway.dev/charts/agentgateway \<br>--namespace agentgateway-system \<br>--version v1.1.0 \<br>--set controller.image.pullPolicy=Always \<br>--set controller.extraEnv.KGW_ENABLE_GATEWAY_API_EXPERIMENTAL_FEATURES=true \<br>--wait

# Run agentgateway in a container<br>$ docker run -p 3000:3000 -p 15000:15000 \<br>-v $(pwd)/config.yaml:/etc/agentgateway/config.yaml \<br>ghcr.io/agentgateway/agentgateway:latest \<br>-f /etc/agentgateway/config.yaml

Quick start<br>Get started with agentgateway in less than 5 minutes — local-first.<br>Standalone &rarr;Kubernetes &rarr;

MCP connectivity guide<br>Connect agentgateway to...