> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cloudeval.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Review IaC before deployment

> Use CloudEval AI as a pre-deployment review layer for ARM and Bicep-based Azure infrastructure.

This workflow is for teams that want faster feedback before infrastructure hits a live environment.

## Best for

* Design review before release
* Pull request review support
* Architecture and cost checks during planning
* Shared review between engineers and stakeholders who do not want to parse raw template JSON

## Workflow

1. Choose the IaC source shape:
   * single ARM JSON template
   * ARM JSON plus parameters file
   * multi-file workspace with linked or nested ARM templates
2. For a multi-file workspace, add `.cloudeval/config.yaml` and set `stacks[].entry` to the visualization source.
3. Create a project from that IaC source.
4. Run architecture and cost reports.
5. Review the findings before the deployment decision.

<Frame caption="Imported infrastructure code can be reviewed beside the project files, report outputs, and generated ARM JSON outline.">
  <img src="https://mintcdn.com/ganakailabs-db727e50/431cYuJMCQHimeRQ/assets/images/workspace/code-editor-arm-template.png?fit=max&auto=format&n=431cYuJMCQHimeRQ&q=85&s=3676ae3ff872f54e6e0c8a171b96b874" alt="CloudEval workspace code editor showing an imported ARM template, file explorer, report files, and outline" width="2160" height="1060" data-path="assets/images/workspace/code-editor-arm-template.png" />
</Frame>

<Frame caption="Code review can stay grounded in a selected template range, file path, line numbers, and the assistant response.">
  <img src="https://mintcdn.com/ganakailabs-db727e50/431cYuJMCQHimeRQ/assets/images/workspace/code-with-ai-review.png?fit=max&auto=format&n=431cYuJMCQHimeRQ&q=85&s=f7c510e14e00b316e71c73a8b22484d1" alt="CloudEval code editor with selected ARM JSON lines and an AI assistant response explaining the selected resource" width="3024" height="1650" data-path="assets/images/workspace/code-with-ai-review.png" />
</Frame>

## Why teams like this path

* It avoids credential setup.
* It keeps code and findings in one workspace.
* It keeps linked ARM files separate in the code view while using a resolved bundle for diagrams and reports.
* It makes pre-deployment review easier for people outside the original authoring context.

## What to look for

* High-severity architecture issues
* Unexpected dependencies or topology shape
* Cost signals that should change SKU or resource choices before rollout

## Next step

If you have not imported a template yet, start with [Import an ARM or Bicep template](/quickstart/import-an-arm-or-bicep-template).