> This page location: /docs.md

> For the complete MongoDB documentation index, see www.mongodb.com/docs/llms.txt

# MongoDB Atlas Agent Engine

MongoDB Atlas Agent Engine is in Public Preview and intended for evaluation and prototyping purposes only. We do not recommend production deployments during preview and expect to communicate breaking changes with you periodically. To learn more about preview feature considerations, see [Preview Features](https://www.mongodb.com/docs/preview-features/).

The MongoDB Atlas Agent Engine is the enterprise-grade platform for developing, deploying, and managing the full lifecycle of AI agents. The Atlas Agent Engine replaces the need for multiple vendors such as vector databases, orchestration platforms, runtime, and synchronization pipelines. Sandboxing and agent identity management are built into the platform so that regulated workloads stay isolated and auditable on every run. You can apply the full stack or one layer of the platform, and you can run on your Atlas clusters.

![Atlas Agent Engine flowchart](/images/AP_agentic_landing_graphic.png)

**Important: Public Preview Limitations**

The MongoDB Atlas Agent Engine is in Public Preview. MongoDB does not recommend that you run production workloads on the platform, and does not provide service-level agreements (SLAs) or service-level objectives (SLOs) for platform availability during Public Preview.

To review all limitations that apply during Public Preview, see [Limitations.](https://www.mongodb.com/docs/agentengine/reference/limitations.md#std-label-agent-engine-limitations)

## How the Atlas Agent Engine is Unique

- **Open and flexible**: Deploy any model or framework across any cloud or on-premises environment. Build with open-source frameworks like LangGraph while MongoDB handles scaling, provisioning, state management, and operational complexity.

- **Strong security, governance, and isolation at scale**: Observe, govern, and sandbox agentic workflows with complete visibility and cost controls. Maintain compliance with GDPR, SOC 2, HIPAA, and industry-specific requirements.

- **Advanced Agentic Memory and State for more efficient and accurate agents**: Automatically manage session state and long-term memory by using integrated vector storage by MongoDB.

- **Designed for enterprise integration**: Connect legacy and third-party tools by using open standards and real-time vector indexing.

**Important: Public Preview Limitations**

The MongoDB Atlas Agent Engine is in Public Preview. MongoDB does not recommend that you run production workloads on the platform, and does not provide service-level agreements (SLAs) or service-level objectives (SLOs) for platform availability during Public Preview.

To review all limitations that apply during Public Preview, see [Limitations.](https://www.mongodb.com/docs/agentengine/reference/limitations.md#std-label-agent-engine-limitations)

## Start Building

You can use the following resources to start building with the MongoDB Atlas Agent Engine:

Learn how to create and deploy a Hello World agent on the Atlas Agent Engine.

Learn how to run an existing agent on the platform.

Enable and configure memory to give your agent long-term context.

## Limitations

Review the scaling, build, and API stability limitations that apply during Public Preview in the [Limitations](https://www.mongodb.com/docs/agentengine/reference/limitations.md#std-label-agent-engine-limitations) section.

## Responsibilities

Atlas Agent Engine users are responsible for the behavior and actions of their agents. Users are responsible for ensuring that their agents comply with MongoDB's [Acceptable Use Policy](https://www.mongodb.com/legal/acceptable-use-policy) and [Terms of Service](https://www.mongodb.com/legal/terms-and-conditions/cloud). Use the [Agent Engine Policy Engine](https://www.mongodb.com/docs/agentengine/manage/governance/policy-engine.md#std-label-agent-engine-policy-engine) to manage and control agent execution.
