Connect Amazon S3 with Blob0
Connect an Amazon S3 bucket to Blob0 on your Mac. Match the endpoint and region, enter your storage credentials, test access, and resolve common connection errors.
Have an S3 bucket already? Use its regional endpoint and authorized access keys to browse and manage objects from your Mac. These instructions follow the Blob0 0.2.0 connection form.
Install Blob0 for macOSBefore you connect
- Find the exact bucket name and its AWS region in the S3 console. This guide uses the illustrative bucket example-documents in us-west-2 (Oregon).
- Obtain an Access Key ID and Secret Access Key authorized for this bucket from your AWS administrator. Listing, reading, uploading, and deleting each need the corresponding permissions.
- Keep the bucket private if that suits your workflow. A public website endpoint or custom domain is not needed to connect through the S3 API.
Blob0 currently accepts an access key pair. Its connection form has no session token field or AWS SSO sign-in. If your organization requires temporary STS credentials or SSO, use an approved client that supports that authentication flow.
Connection fields
On first launch, use the connection form. In an existing workspace, open Settings → Connections → Add connection. Choose Amazon S3 under Provider before filling in the remaining fields.
For a bucket in us-west-2, use the following example. Replace the bucket name and keys with your own values; the example contains no working credentials.
| Field | What to enter |
|---|---|
| Provider | Amazon S3 |
| Connection name | A local label, for example Work documents |
| S3 API endpoint | https://s3.us-west-2.amazonaws.com |
| Bucket | example-documents — the bucket name only, without s3:// or a folder path |
| Region | us-west-2 — the actual bucket region, not the R2 value auto |
| Access Key ID | Your authorized AWS access key ID |
| Secret Access Key | The secret for the same access key; stored in the macOS Keychain |
| Public / custom domain | Optional. Leave empty for a private bucket |
The Amazon S3 preset uses virtual-hosted addressing and does not show a path-style switch. The endpoint above is the regional API host; Blob0 adds the bucket name when making requests. A different region needs its matching endpoint, so review both values after changing providers.
Test, save, and check your first file
Select Test connection
A successful Connected response checks listing access. A failure leaves the profile unsaved so you can correct the fields. An empty root listing can be a valid result.
Select Add connection
Save the profile and connect. In Settings, switch to the saved connection if it is not already active. You should see the bucket’s root objects and folder prefixes.
Check the permissions you need
If authorized, upload a small disposable file to a test prefix, wait for its transfer to finish, download it, and compare its contents. Delete only that test object when finished. A listing test alone does not prove upload, download, or delete permission.
If the connection fails
- AccessDenied
- Ask the bucket administrator to check the key’s IAM permissions and the bucket policy for the attempted operation. Encryption with AWS KMS can require additional permissions. Making the bucket public is not a substitute for authorized access.
- SignatureDoesNotMatch
- Check that the access key and secret belong together, remove accidental whitespace, and verify the bucket region and API endpoint. Also check your Mac’s clock.
- Wrong region or redirect
- Match Region to the bucket’s actual AWS region and use that region’s S3 API endpoint. Do not copy the Cloudflare R2 auto region into an AWS connection.
- Cannot reach the endpoint
- Use an HTTPS S3 API endpoint rather than a website URL, then check your network, DNS, proxy, or firewall. For nonstandard AWS partitions, obtain the correct endpoint from AWS documentation.
What this connection is for
Use Blob0 to work with objects in a connected bucket: browse, transfer, preview, organize, and share. Bucket creation, IAM administration, storage billing, and AWS SSO configuration remain in your AWS tools. Rename and move operate within the active bucket; they are not cross-bucket migration.