How Kagenti ADK simplifies production AI agent management

Kagenti ADK is an open source toolkit that handles the operational complexity of running AI agents in production. It implements the Linux Foundation's Agent2Agent (A2A) protocol, providing automatic agent card generation, JSON-RPC 2.0 message handling, and mutual TLS via Istio Ambient Mesh. Beyond A2A compliance, it ships runtime services injected via dependency injection: an LLM proxy supporting 15+ providers, PostgreSQL for session state, pgvector for RAG, MCP connectors, Keycloak-based authentication with SPIFFE workload identity, OpenTelemetry-based LLM tracing, and S3-compatible file storage. A CLI handles scaffolding, local runs, and Kubernetes/OpenShift deployments. Pre-built reference agents demonstrate integration patterns including two-stage retrieval, trajectory tracing, and citation rendering.