AWS S3 integration

Overview

Amazon S3 (Simple Storage Service) is AWS's object-storage service – buckets of objects (files) with versioning, tagging, lifecycle rules, and global durability. The Automize AWS S3 connector covers the operations you actually need in automation: bucket + object CRUD, listing, copying, multi-delete, tagging, and object metadata.

• Regions: S3 is regional. Each bucket lives in one AWS region (us-east-1, eu-west-1, etc.). The connector picks the region from the credentials configured on the connection – if you span regions, use separate connections.
• Plans: Pay-as-you-go AWS. No plan gating. Pricing scales with storage GB + requests + data transfer.

Setting up the connection

Auth type: aws_sigv4.

1. In AWS Console → IAM → Users → Add user → Programmatic access.
2. Attach a policy with the S3 actions you need (s3:GetObject, s3:PutObject, s3:ListBucket, etc.) scoped to the relevant buckets.
3. Create an access key. Copy Access Key ID + Secret Access Key.
4. In Automize, open Settings → Connectors → AWS S3 → Add connection. Paste keys + default region.

Find your credentials at https://us-east-1.console.aws.amazon.com/iam/home.

Rate limits

S3 supports very high request rates – 3,500 PUT/POST/DELETE and 5,500 GET/HEAD per second per prefix. Prefixes auto-scale further. If you hit SlowDown errors, randomise object key prefixes (avoid sequential date stamps as the leading bytes of keys).

Data model

Buckets contain objects. Object keys are flat strings, but
forward-slashes simulate a folder hierarchy in tools.

    Bucket ─── Objects (keyed by 'path/to/file.ext')
           └── Versions (when versioning enabled)
    ACLs + Bucket Policies + IAM (multi-layered authorisation)

Available operations

13 operations available. Click any row to jump to its detail.

Operations

Copy object

Copy an object to another location in AWS S3.

Tips

• Server-side copy – no bytes transit your network. Use for cross-bucket / cross-region (with same account) moves.
• Cross-region copy: source bucket policy + dest bucket policy must allow.

Create bucket

Create a new AWS S3 bucket.

Tips

• bucket name must be globally unique, lowercase, 3-63 chars, DNS-compliant.
• Region picked from connection settings; pass LocationConstraint for non-us-east-1.

Errors

• BucketAlreadyExists – Bucket name taken globally. Pick a more unique name.
• BucketAlreadyOwnedByYou – Already created in another region by you.

Delete bucket

Delete an empty AWS S3 bucket.

Tips

• Bucket must be empty. Use List Objects + Delete Objects first.

Delete Object

Delete an object

Tips

• Soft-delete if versioning enabled (creates delete marker). Hard-delete otherwise.

Bulk Delete Objects

Delete multiple objects in one request (up to 1000)

Tips

• Up to 1,000 keys per call. Returns per-key success/failure.

Get Object

Download an object

Tips

• Returns object bytes + metadata. range param for partial downloads (HTTP Range header).

Get object tags

Get tags on an AWS S3 object.

Tips

• Tags are key-value metadata. Up to 10 tags per object.

Head object

Get metadata for an AWS S3 object without downloading it.

Tips

• Metadata only – content-length, content-type, etag, last-modified. Cheaper than Get when you don't need the bytes.

List Buckets

List all S3 buckets

No input parameters.

Tips

• Returns all buckets owned by the AWS account. No filtering – buckets are global.

List Object Versions

List versions of objects in a bucket

Tips

• Versioning must be enabled. Returns version IDs + delete markers.

List Objects

List objects in a bucket

Tips

• Pages 1,000 keys per call. Pass continuation_token for next page.
• prefix='folder/' to scope. delimiter='/' for folder-style listings.

Put Object

Upload an object

Tips

• Simple upload up to 5 GB. For larger, use multipart upload (separate API not exposed).
• Set Content-Type explicitly. SSE-S3 or SSE-KMS for encryption-at-rest.

Set object tags

Set tags on an AWS S3 object.

Tips

• Replaces all tags (not a merge). Pass empty tags=[] to clear.

FAQ

Why does my List Objects return 1,000 only?

S3 caps each response at 1,000 keys. Pass continuation_token from one response into the next call to paginate.

Multipart upload for big files – how?

Not exposed in this connector. Workaround – split client-side or use the AWS CLI for one-offs. Multipart is needed for files >5 GB; required >100 MB for resilience.

Presigned URLs?

Generate via the AWS SDK (separate; connector doesn't expose). Use to give end users time-limited direct download/upload links without exposing your AWS credentials.

Cross-account access?

Bucket policy + IAM on both sides. The connector uses the credentials configured – for cross-account, configure those credentials with sts:AssumeRole upstream.

Related connectors

• Dropbox
• Box
• Azure-blob
• Google-workspace