What is STRG?
STRG is S3-compatible object storage for businesses, individuals and AI agents. You create buckets in a region, keep objects in them, and reach them however suits the moment: from the web dashboard, from any tool or SDK that speaks the S3 API, from an AI agent through the STRG MCP server - and, for managing access keys from scripts, through the STRG REST API. That makes it a good home for a nightly backup, the uploads of a web app, a static site's assets, or an agent keeping its memory in a bucket. Ease of use comes first: there is no infrastructure to configure and no request pricing to model, and agents are first-class users rather than an afterthought.
Billing
The idea is simple: you pay for the data you store, and nothing else.
| Stored data | Billed |
| API requests | No charge |
| Ingress (uploads) | No charge |
| Egress (downloads) | No charge |
Every stored object version counts as stored data. Buckets are always versioned, so an object you delete stays billed until its older versions are removed - by you, or by a lifecycle policy. Objects under Object Lock are billed until their lock expires.
A card goes on the dashboard's Billing page before your first bucket or access key. In the dashboard's words: "A card is required to prevent abuse: a $1 hold checks it and is released at once, and nothing is charged until there is storage to bill."
Stored data is invoiced and charged to your card. Billing walks through invoices and cards.
Ways to use STRG
Each surface reaches the same buckets and keys, so you can mix them freely - create a bucket in the dashboard, fill it from a script, and let an agent read from it.
| Surface | What it is | Start here |
|---|---|---|
| Dashboard | The web interface at https://app.strg.com, for signing up, creating buckets, browsing and uploading objects, and managing bucket settings, access keys and billing, with the Ask our AI Assistant beside every page | Getting started in the dashboard |
| S3-compatible API | https://s3.<region>.strg.com, for the AWS CLI, the AWS SDKs and any other S3 client, signed with an access key | S3 API |
| MCP server | https://mcp.strg.com/mcp, which lets an AI agent list and create buckets, run S3 operations on buckets and objects, create presigned URLs, and create, list and revoke access keys, and answers questions about STRG from its documentation | Getting started with an AI agent |
| REST API | https://api.strg.com/v1/, a JSON API for listing, looking up, creating and deleting access keys from scripts, CI and tools, authenticated with an access key that holds Manage access keys through the API | REST API |
For help choosing between them, Getting started compares the ways in.
Features
Here is what STRG offers today, each with the page that explains it in full.
| Feature | What you get |
|---|---|
| Buckets and regions | Bucket names are unique across STRG and 3-58 characters long. You pick a region per bucket; a bucket's name and region stay as they were set when it was created. Every bucket is encrypted at rest with AES-256, and its page in the dashboard shows its S3 endpoint |
| Bucket Protection | Every bucket is Protected (private) unless you turn protection off, which keeps it reachable only with your access keys. With protection off - handy for a public website or shared downloads - anyone can read, but not list, its objects without credentials at https://<bucket>.s3.<region>.strg.com/<key> - https://s3.<region>.strg.com/<bucket>/<key> for a bucket name with a dot - and the dashboard offers Copy public URL for each object. An object whose own Cache-Control allows a shared cache is cached for that long, at most 5 minutes - so an overwrite, a delete, turning protection back on or a CORS change can take up to 5 minutes to reach anonymous readers |
| Objects and versioning | The dashboard lets you browse, upload, download and delete objects and folders. Buckets are always versioned, so nothing is lost by accident: a delete leaves a delete marker, and earlier versions can be listed, restored and undeleted |
| Access keys | S3 access keys scoped to all buckets or to specific ones, optionally to a key prefix, with the permissions you choose. A key can also be enabled for the MCP server, or for managing access keys through the REST API |
| CORS rules | Per-bucket CORS configuration in S3's own JSON format, so an existing AWS configuration pastes straight in. The rules apply to browser requests to the bucket on the S3 endpoints - which is what a web app uploading straight from the browser needs |
| Lifecycle policies | Rules per prefix that, after a number of days, expire current versions, permanently delete previous versions and abort incomplete multipart uploads - a tidy way to keep old backups and their cost in check |
| Object Lock | Write-once-read-many retention in governance or compliance mode, with a default retention per bucket and legal holds per object. Enabling it on a bucket is permanent |
| Ask our AI Assistant | A chat panel beside every dashboard page that does storage work for you, reads the page you are on and fills in its forms for you to review and submit. It asks you to confirm anything that cannot be undone |
| Two-factor authentication and account settings | Account settings, opened from the avatar menu, is where you can change your email address, turn on two-factor authentication with an authenticator app, or close your account |
| S3 API | The common S3 operations for buckets, objects, multipart uploads, versions, CORS, lifecycle, default encryption and Object Lock. The page lists every supported operation and how each differs from AWS S3 |
Regions
A bucket lives in the region you create it in. These are the regions available today:
| Region | Name | S3 endpoint |
|---|---|---|
us-east-1 | US East | https://s3.us-east-1.strg.com |
eu-central-1 | EU Central | https://s3.eu-central-1.strg.com |
Each regional endpoint serves that region's buckets, so a client is configured with the endpoint and region of the bucket it talks to. S3 API covers the addressing styles and the differences from AWS S3.