> [!NOTE]
> Features mentioned on this page, such as AI Gateway and Agent Firewall,
> require the [AI Governance Add-On](https://coder.com/docs/ai-coder/ai-governance.md). As of Coder v2.32,
> deployments without the add-on will not be able to access these features.

As the AI landscape is evolving, we are working to ensure Coder remains a secure
platform for running AI agents just as it is for other cloud development
environments.

## Use Trusted Models

Most agents can be configured to either use a local LLM (e.g. llama3), an agent
proxy (e.g. OpenRouter), or a Cloud-Provided LLM (e.g. AWS Bedrock). Research
which models you are comfortable with and configure your Coder templates to use
those.

## Set up Firewalls and Proxies

Many enterprises run Coder workspaces behind a firewall or a proxy to prevent
threats or bad actors. These same protections can be used to ensure AI agents do
not access or upload sensitive information.

## Separate API keys and scopes for agents

Many agents require API keys to access external services. It is recommended to
create a separate API key for your agent with the minimum permissions required.
This will likely involve editing your template for Agents to set different
scopes or tokens from the standard one.

Additional guidance and tooling is coming in future releases of Coder.

## Set Up Agent Firewall

Agent Firewall is a process-level firewall that lets you restrict and
audit what AI agents can access within Coder workspaces. To learn more about
this feature, see [Agent Firewall](https://coder.com/docs/ai-coder/agent-firewall.md).