Skip to main content
Open Managed Agents offers two ways to build, each suited to different use cases: Open Managed Agents provides the harness and infrastructure for running autonomous agents. Instead of building your own agent loop, tool execution, and runtime, you get an environment you manage where the agent can read files, run commands, browse the web, and run code securely. The harness supports built-in prompt caching, compaction, and other performance optimizations for high-quality, efficient agent outputs.
OMA is open source and self-hosted. The operator manages databases, object storage, model credentials, and Environment Runners. “Cloud environment” in these docs means an E2B or compatible runner configured by your OMA deployment.

Quickstart

Create your first agent session

Start a session

Create a session and send your first event

Reference

Event types, rate limits, CLI flags, and other lookup tables

​
Core concepts

Open Managed Agents is built around four concepts:

​
How it works

1

Create an agent

Define the model, system prompt, tools, MCP servers, and skills. Create the agent once and reference it by ID across sessions.
2

Create an environment

Configure where the agent runs: a cloud sandbox supplied by an Environment Runner, or a self-hosted sandbox on your own infrastructure.
3

Start a session

Launch a session that references your agent and environment configuration.
4

Send events and stream responses

Send user messages as events. The agent autonomously runs tools and streams back results through server-sent events (SSE). Event history is persisted server-side and can be fetched in full.
5

Steer or interrupt

Send additional user events to guide the agent mid-execution, or interrupt it to change direction.

​
When to use Open Managed Agents

Open Managed Agents is best for workloads that need:
  • Long-running execution: Tasks that run for minutes or hours with multiple tool calls
  • Cloud infrastructure: Secure sandboxes with pre-installed packages and network access
  • Self-hosted execution: Sandboxes on infrastructure you control for compliance or data-residency requirements
  • Minimal infrastructure: No need to build your own agent loop, sandbox, or tool execution layer
  • Stateful sessions: Persistent filesystems and conversation history across multiple interactions
  • Scheduled execution: Recurring agent runs on a cron schedule through scheduled deployments

​
Supported tools

Open Managed Agents provides a set of built-in tools:
  • Bash: Run shell commands in the sandbox
  • File operations: Read, write, edit, glob, and grep files in the sandbox
  • Web search and fetch: Search the web and retrieve content from URLs
  • MCP servers: Connect to external tool providers
See Tools for the full list and configuration options.

​
Beta access

Open Managed Agents is in beta. All Managed Agents endpoints require the managed-agents-2026-04-01 beta header. Compatible SDKs set the beta header automatically. Behaviors may be refined between releases to improve outputs.
To get started, you need:
  1. A reachable OMA deployment and workspace API key
  2. The managed-agents-2026-04-01 beta header on all requests
  3. A configured environment that can start sessions
MCP tunnels and dreaming remain planned capabilities. The current release does not yet expose the complete Dreams API; see the implementation-status warning at the top of Dreams. Open Managed Agents is stateful by design: sessions are long-running, resume cleanly after pauses, and store conversation history, sandbox state, and outputs in your deployment. Retention, encryption, and compliance boundaries are controlled by the operator. You can delete sessions, and separately delete any uploaded files, at any time through the API.