> For the complete documentation index, see [llms.txt](https://doc.wearepatchworks.com/product-documentation/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.wearepatchworks.com/product-documentation/developer-hub/custom-scripting/deploying-scripts-and-managing-versions.md).

# Deploying scripts & managing versions

### Introduction

Scripts are versioned, and saving is separate from deploying:

* **Saving** a version stores your work. You can [save](https://doc.wearepatchworks.com/product-documentation/developer-hub/custom-scripting/creating-and-testing-custom-scripts/creating-a-script-manually#saving-your-work) and [test](https://doc.wearepatchworks.com/product-documentation/developer-hub/custom-scripting/creating-and-testing-custom-scripts/testing-scripts) as often as you like, without affecting anything that's live.
* **Deploying** a version releases it. When you deploy, you can also choose which process flows using the script should move onto the new version — so rolling out a change everywhere is a single action.

### Need to know

* A script can hold up to **10 versions**. Once the limit is reached, no more versions can be added until you delete enough old ones to bring the count back down.
* Deploying a different version is also how you **roll back** — if a change causes problems, simply deploy the previous version again.
* Every change and deployment is recorded in the platform logs, so you can always see who changed a script, when it was deployed, and which flows were updated.

### Deploying a version

Follow the steps below:

**Step 1**\
Open your script and go to `versions` > `deploy`.

**Step 2**\
Choose the version to release from the `version to deploy` dropdown. The selected version is shown read-only, so you can confirm exactly what you're about to release — if the version you've selected is already live, a confirmation banner is displayed:

<div data-with-frame="true"><figure><img src="https://2440044887-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FLYNcUBVQwSkOMG6KjZfz%2Fuploads%2Flpw1Sl13BGsJdd1reEP6%2FDeploy%20Screen%20.png?alt=media&amp;token=ea24779c-1be1-4efd-8319-5cc0f578eb05" alt=""><figcaption></figcaption></figure></div>

**Step 3**\
Deploy the version. From this point, it's the live version of the script.

### Updating flows that use this script

The `flows using this script` panel (right) gives you full visibility of where the script is used — every process flow is listed, together with the version it currently uses.

To move flows onto the version you're deploying:

**Step 1**\
Tick the flows you want to update. The change each flow will receive is shown alongside (for example, `v1 → v2`) — flows already on the selected version are flagged, so there's nothing to do for those.

**Step 2**\
Click the `update` button (for example, `update 1 flow`) — each selected flow is updated to use the new version.

{% hint style="warning" %}
Only a flow's **draft** can be changed. A live flow keeps running its currently deployed script version until you deploy the flow itself — so updating a script version never changes live behaviour behind your back.&#x20;
{% endhint %}

### Pinned versions & rollbacks

Wherever a script is attached (a script shape, map shape, connector endpoint, etc.), it references a specific script version — as shown in the `currently uses` column on the deploy page. This means:

* Deploying a new script version doesn't force it on anyone — flows move onto the new version when you update them (above) and deploy them.
* To roll back, deploy an earlier version and update the relevant flows onto it in exactly the same way.

If you'd rather a given usage always picks up your newest release, select the `use latest version` option where the script is attached — those usages follow whichever version is deployed, with no need to update them each time.

### Audit trail

Script changes and deployments are captured in the platform logs — including who edited a script, when versions were deployed, and which flows were updated to use them. This gives you a complete history for troubleshooting and change control.