Delete by Prefix
Deletes every object whose key starts with a prefix, such as a whole folder, from an Amazon S3 bucket.
Delete by Prefix removes every object in an Amazon S3 bucket whose key starts with the text you give. Use it to empty a folder, for example temp/2026-11-10/.
Before you start
The AWS S3 actions don't use a stored credential. Each action has the same four connection fields:
| Field | Type | Required | What to put |
|---|---|---|---|
| Access Key | Text | Yes | The access key ID of an AWS IAM user, for example AKIAโฆ. |
| Secret Key | Text | Yes | The secret access key that belongs to that access key. |
| Region | Text | Yes | The AWS region of the bucket, as its code, for example eu-central-1. |
| Bucket | Text | Yes | The bucket name, for example my-company-invoices. |
The IAM user needs S3 permissions on the bucket for the operation you run. The values you type are visible to everyone who can edit the process, so prefer an IAM user limited to the buckets the process needs.
Add the action
In the Process Designer toolbox, open AWS S3 and pick Delete by Prefix (or type Delete by Prefix in the search box).
Inputs and outputs
| Field | Direction | Type | Required | What to put / what you get |
|---|---|---|---|---|
| Access Key, Secret Key, Region, Bucket | Input | Text | Yes | The connection fields described above. |
| Prefix | Input | Text | Yes | The start of the keys to delete, for example temp/2026-11-10/. |
| Result | Output | Boolean | No | true if the objects were deleted, false if nothing matched. |
Example
Prefix temp/2026-11-10/ โ every object under temp/2026-11-10/ is removed and Result is true.
The prefix matches the start of the key, not a folder name. reports matches reports/2025.csv and reports-archive/old.csv. End the prefix with / to delete only that folder. The deletion can't be undone.
Good to know
- An empty Prefix is refused with "Missing prefix.", so a missing value can't empty the whole bucket.
- When no object matches, Result is
falseand the step doesn't fail. - One run handles at most the first 1,000 matching objects (one S3 listing page). For larger folders, run the action again until Result is
false. - If a connection field is empty, the step fails with "Invalid connection details."
- Common errors from S3: "Access forbidden! Ensure the connection details are valid." (wrong keys or missing permission), "Unable to connect! Ensure the provided region matches the bucket one." (wrong Region), and "Unrecognized Amazon Region: โฆ" (a region code that doesn't exist).