Documentation
Features#
Everything works with a single helm install. No add-ons, no extra databases, no waiting for enterprise.
Agent lifecycle via CRDsDefine, version, and roll out agents with kubectl and GitOps — the same workflow as every other workload.
Multi-runtime supportGo and Python ADK runtimes. Pick the language that fits, or mix both in the same cluster.
BYO frameworksLangGraph, CrewAI, Google ADK, or your own — bring any agent framework and kagent orchestrates it.
Long-term memoryPersistent vector-backed memory across sessions. Agents remember context, not just the last prompt.
Human-in-the-loopTool approval gates, agent-initiated questions, and cascading HITL — humans stay in control.
Agent-to-Agent (A2A)Agents discover and invoke each other. Compose multi-agent workflows with first-class delegation.
Skills from GitLoad markdown knowledge from Git repos at startup. Agents learn your runbooks, ADRs, and internal docs.
Prompt templatesReusable prompt fragments stored as ConfigMaps. DRY your system prompts across agents.
Context compactionAuto-summarization of long conversation histories. Agents stay coherent without blowing token budgets.
Sandbox & securityAgent sandboxing, RBAC, and security hardening out of the box. Run untrusted code safely.
Full observabilityOTel tracing, Prometheus metrics, structured logs. See every prompt, every tool call, every token.
Postgres storageProduction-grade Postgres-backed storage with reviewable migrations. No proprietary database lock-in.
Kagent Lab: Discover kagent and kmcp
Free, on‑demand lab: build custom AI agents with kagent and integrate tools via kmcp on Kubernetes.