01 / behavior
Agent
The model, system prompt, tools, MCP servers, and skills that define how Codex behaves.
Private early access
Join the waitlist for managed Codex agents.
Codex Managed Agents is rolling out privately. Join the waitlist to share your workflow, and we will reach out as we open early access for teams that need secure tool execution, cloud environments, and long-running Codex sessions.
Secure command, file, and web execution
Cloud environments with persistent session state
Private early access for real agent workloads
What you are signing up for
Private access still starts with four runtime primitives.
Model + tools
Agent
Define Codex behavior once.
Container template
Environment
Bring packages, files, and access rules.
Execution runtime
Session
Combine the agent and environment for a live task.
App I/O
Events
Send instructions in and stream tool/status updates out.
Core concepts
The platform still maps cleanly to four runtime objects.
Agent, Environment, Session, and Events remain the core vocabulary of the system. Even in private rollout, they are the fastest way to understand what the product is designed to run.
02 / runtime
Environment
A configured container template with pre-installed packages, mounted files, and network rules.
03 / execution
Session
A running agent instance inside an environment, working on a concrete task and producing outputs.
04 / control plane
Events
Messages flowing between your application and the agent, including user turns, tool results, and status updates.
When access opens
Invited teams follow the same short path from setup to execution.
Once you have access, the workflow stays intentionally small: define Codex behavior, choose the runtime, launch a session, and keep steering it through events while the infrastructure stays managed.
01
Define an agent
Set the model, prompt, tools, MCP servers, and skills once, then reuse the agent definition across tasks.
02
Provision the runtime
Choose the cloud container template with the packages, files, and access controls your workload needs.
03
Launch a session
Start an execution that binds the chosen agent and environment for a specific task.
04
Stream events and outputs
Send user input as events while Codex streams back status updates, tool calls, and outputs over SSE.
05
Guide work in flight
Interrupt, redirect, or refine the session without rebuilding the whole runtime loop.
Who should join
Best for teams planning durable agent workloads.
Codex Managed Agents is not just a UI wrapper around a model. It is a runtime layer for workloads that need time, tools, isolation, and a memory of what already happened, which is exactly what the waitlist helps us prioritize.
Long-running execution
Handle workloads that span minutes or hours and require multiple coordinated tool calls.
Cloud infrastructure
Run inside managed containers with pre-installed packages, mounted files, and explicit network access.
Minimal infrastructure
Skip building your own agent loop, sandbox, and tool orchestration layer from scratch.
Stateful sessions
Keep a persistent file system and conversation history across multi-turn tasks and retries.
Waitlist open
Tell us what you want to run with Codex Managed Agents.
We are onboarding teams with real agent workflows first. Share your use case through the waitlist, then browse the docs while you decide how Codex Managed Agents could fit into your stack.