Agentgateway
An open-source proxy for agentic AI that provides connectivity, security, observability and governance for agent-to-LLM, agent-to-tool and agent-to-agent communication, built on the MCP and A2A protocols.
Agentgateway's LLM gateway routes traffic to major providers such as OpenAI, Anthropic, Gemini and Bedrock through a unified OpenAI-compatible API with budget and spend controls, prompt enrichment, load balancing and failover. Its MCP gateway connects LLMs to tools and external data with tool federation across stdio, HTTP, SSE and Streamable HTTP transports, OpenAPI integration and OAuth authentication, and its A2A gateway enables secure agent-to-agent communication with capability discovery, modality negotiation and task collaboration.
Inference routing directs traffic to self-hosted models using Kubernetes Inference Gateway extensions, based on GPU utilization, KV cache, LoRA adapters and queue depth; multi-layered guardrails combine regex, OpenAI moderation, AWS Bedrock Guardrails, Google Model Armor and custom webhooks; and security features cover JWT, API-key and OAuth authentication, CEL-based RBAC, rate limiting, TLS, and OpenTelemetry metrics, logs and tracing.
It is written in Rust, deploys standalone with flat YAML configuration or on Kubernetes with a built-in controller and the Gateway API, and is licensed under Apache-2.0.