Guides
File hosting guides for agents and automation.
Use these guides to publish agent deliverables, automate uploads, and choose the correct hosting workflow.
Getting started
-
How AI agents can host files without an email address
One POST request creates an account and returns the API key. A $1 payment grants 1 GiB, and the account can then publish files at durable public URLs.
Automation workflows
-
Turn a CI report into a durable public URL
A CI job can publish its PDF report or build log with one curl call and hand the resulting URL to a ticket, a chat message, or the next pipeline stage.
Engineering
-
Content-addressed file hosting: same bytes, same URL
DurableFile URLs contain the SHA-256 of the file bytes. Identical bytes map to one stored object and one URL, so retried uploads are free and idempotent.
-
Temporary links that can come back: expiry, unpublish, republish
A link can expire on schedule or be hidden manually, keep its bytes while unpublished, and return at the identical URL — only a delete frees quota.
-
Why an upload API returns HTTP 402 and how an agent should handle it
HTTP 402 is a recoverable state, not a failure. The response body carries a fix instruction, a Stripe checkout URL for humans, and machine-payable stablecoin options for agents.
Choosing a host
-
Where should an AI agent put a generated PDF?
Pastebins choke on binary files, presigned URLs expire and need cloud credentials, data URIs bloat and break — a durable public URL is the missing default for agent deliverables.