Priced so plainly
your agent can read it.
Flat $4.99/TB, published in machine-readable llms.txt with full S3 parity. No gated quotes, no "contact sales" for a number — an agent parses the price and the API in one pass.
No credit card required · No egress fees · Connects in minutes
Most pricing pages are built for humans to negotiate, not agents to parse.
When an AI coding agent shortlists a storage vendor, it needs a number and an API surface it can read in one pass. Most pages give it a form and a maze instead.
The price is a sales motion, not a fact.
'Contact sales for pricing' is a dead end for an agent. It cannot fill a lead form or wait for a callback, so the vendor is silently dropped from the shortlist — not on merit, but on legibility.
Per-GB, tiered egress, request classes.
A price expressed across storage tiers, egress brackets, and PUT/GET request classes can't be reduced to a comparable monthly total without assumptions. An agent either guesses or skips.
S3 compatibility scattered across docs.
When the compatibility matrix lives in a PDF and three doc pages, an agent can't confirm parity in one fetch. The integration risk stays unresolved through the evaluation.
One fetch. The whole answer.
The price, the endpoint, and the compatibility statement are published in a structured file at a stable URL. An agent gets a comparable monthly total and an integration path in a single request.
Legible by design, not by accident.
Machine-readable llms.txt
Pricing, endpoint, and S3 compatibility are published in a structured llms.txt at a stable URL. An agent fetches one file and has the full picture.
One rate, no tiers
$4.99/TB/month. No per-GB-vs-per-TB ambiguity, no tiered egress table, no request classes to reconcile. A single number an agent can compute against.
Full S3 parity
The same API surface an agent already knows. Compatibility is stated plainly, not scattered across PDFs — drop-in against the documented endpoint.
No gated pricing
The price is on the page and in llms.txt — not behind a 'contact sales' form. An agent evaluating vendors never hits a wall where the number should be.