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

# Release management

# Release Management

Liberate provides release management capabilities that enable teams to version workflows, create release artifacts, and deploy to QA or Production environments with confidence.

## Overview

Release management in Liberate follows a simple flow:

```
Edit Workflow → Revisions (auto-saved) → Create Release → Deploy to QA → Deploy to Production
```

| Concept | Description |
| - | - |
| **Revision** | An auto-saved snapshot of your workflow. Every change creates a new revision. |
| **Release** | A tagged version of a revision, ready for deployment. Includes release notes. |
| **Deployment** | Pushing a release to QA or Production environment. |
| **Rollback** | Reverting to a previous release if issues occur. |

***

## Revisions

Revisions are automatic snapshots of your workflow saved whenever you make changes.

### How Revisions Work

* **Auto-saved**: Liberate automatically saves a revision whenever there's a difference in the workflow
* **Tracked history**: Every change is recorded, allowing you to see the full evolution of your workflow
* **Recoverable**: You can view, compare, or restore any previous revision

### Viewing Revision History

1. Open your workflow in the canvas
2. Click **Version History** at the top of the canvas
3. Browse the list of all saved revisions

Each revision shows:

* Timestamp of when it was saved
* Who made the change (if applicable)
* Summary of modifications

> [!TIP]
>
> Use revision history to understand how a workflow evolved or to recover from accidental changes.

***

## Releases

Releases are tagged versions of your workflow that are ready for deployment. Unlike revisions (which are automatic), releases are intentionally created when you're ready to deploy.

### Creating a Release

**From the current version:**

1. Click the top menu in the workflow canvas
2. Select **Create Release**
3. Add release notes describing the changes
4. Click **Create**

**From a previous revision:**

1. Open **Version History**
2. Find the revision you want to release
3. Click on the revision and select **Create Release**
4. Add release notes
5. Click **Create**

### Release Notes

Always include release notes with your releases. Good release notes should describe:

* What changed (new features, bug fixes, improvements)
* Why it changed (business requirement, bug report, optimization)
* Any dependencies or considerations for deployment

**Example release notes:**

```
v2.3.0 - FNOL Workflow Update

Changes:
- Added support for flood claim types
- Fixed policy number validation for HO-prefix policies
- Improved error handling for Snapsheet API timeouts

Dependencies:
- Requires Snapsheet connection v1.2 or higher
```

***

## Deploying Releases

The **Release Tab** on the right panel of the workflow interface manages deployments.

### Deployment Targets

| Environment | Purpose | When to Deploy |
| - | - | - |
| **QA** | Testing and validation | After creating a release, before production |
| **Production** | Live customer traffic | After QA validation is complete |

### How to Deploy

1. Open the **Release Tab** in the right panel
2. Select the release you want to deploy
3. Choose the target environment (QA or Production)
4. Confirm the deployment
5. Monitor for any errors or issues

> [!WARNING]
>
> Always deploy to QA first and validate thoroughly before deploying to Production.

### Deployment Checklist

Before deploying to Production, verify:

* [ ] Workflow tested in Preview mode with sample data
* [ ] Release deployed and validated in QA environment
* [ ] Connections configured for Production (different credentials than QA)
* [ ] Stakeholders notified of upcoming changes
* [ ] Rollback plan in place if issues occur

***

## Rollback

If a deployment causes issues, you can quickly rollback to a previous stable release.

### When to Rollback

* API errors after deployment
* Unexpected workflow behavior
* Data transformation issues
* Integration failures

### How to Rollback

1. Open the **Release Tab** in the right panel
2. Find the previous known-good release
3. Select **Deploy** for that release
4. Choose the environment to rollback (QA or Production)
5. Confirm the rollback

> [!TIP]
>
> Keep track of which release is currently deployed to each environment. This makes rollbacks faster when issues occur.

### Rollback Best Practices

* **Document issues**: Note what went wrong before rolling back
* **Communicate**: Inform stakeholders about the rollback
* **Investigate**: After rolling back, debug the issue in Preview before re-deploying
* **Test thoroughly**: Ensure the fix works in QA before attempting Production again

***

## Best Practices

### Versioning Strategy

Use semantic versioning for releases:

| Version | When to Use |
| - | - |
| **Major (v2.0.0)** | Breaking changes, major rewrites |
| **Minor (v1.3.0)** | New features, significant improvements |
| **Patch (v1.2.1)** | Bug fixes, minor adjustments |

### Release Workflow

```
1. Develop in Preview
       ↓
2. Test with sample data
       ↓
3. Create Release with notes
       ↓
4. Deploy to QA
       ↓
5. Stakeholder validation
       ↓
6. Deploy to Production
       ↓
7. Monitor for issues
```

### Environment Parity

Keep QA and Production as similar as possible:

* Same workflow logic
* Same connection configurations (different credentials)
* Same retry strategies and error handling

This ensures QA testing accurately predicts Production behavior.

***

## Next Steps

> [card] [Preview Mode](https://docs.liberate.ai/workflows/preview-mode)
>
> Test workflows before creating releases

> [card] [Connections](https://docs.liberate.ai/workflows/concepts/connections.md)
>
> Configure environment-specific credentials

> [card] [Error Handling](https://docs.liberate.ai/workflows/nodes/error-handling)
>
> Build robust error handling into workflows

> [card] [Workflow Quickstart](https://docs.liberate.ai/workflows/quickstart.md)
>
> Learn the basics of building workflows


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.