AgentOven Documentation
Everything you need to install, configure, and operate AgentOven — from your first agent to a production fleet.
What is AgentOven?
AgentOven is an open-source enterprise control plane for AI agents — think "Kubernetes for AI Agents." It provides a unified platform to register, deploy, orchestrate, and observe autonomous agents regardless of the underlying framework (LangChain, CrewAI, AutoGen, custom).
Why AgentOven?
Framework-Agnostic
Works with any agent framework. Register LangChain, CrewAI, AutoGen, or custom agents side-by-side.
A2A Protocol Native
Every agent gets a standards-compliant Agent-to-Agent endpoint automatically — agents discover and talk to each other.
Full Observability
OpenTelemetry traces, Prometheus metrics, structured logs — see exactly what every agent is doing.
Workflow Engine
Chain agents into DAG-based Recipes with fan-out, fan-in, conditional branching, and retry policies.
Enterprise Ready
Multi-tenant kitchens, RBAC, audit logging, secrets management, and horizontal scaling.
Quick Example
Get an agent running in under 60 seconds:
terminal
$ cargo install agentoven-cli $ agentoven init my-kitchen 🏺 Created agentoven.toml $ agentoven agent register --name research-bot --framework langchain ✓ Agent 'research-bot' registered $ agentoven agent bake research-bot 🔥 Baking agent 'research-bot'... done! ✓ Agent is hot and ready at /agents/research-bot/a2a
Architecture at a Glance
AgentOven is built as a four-layer stack:
CLI & SDKs
Rust CLI, Python / TypeScript / Rust SDKs — the developer interface.
Control Plane
Go service — agent registry, recipe engine, task router, auth.
Protocol Layer
A2A + MCP — agent-to-agent communication and tool integration.
Infrastructure
PostgreSQL, Redis, OpenTelemetry Collector — the backing services.
Ready to get started? Head to the Quick Start guide, or explore Core Concepts to learn the kitchen vocabulary.