dotsweep vs Fastly Domain Research API
The product Domainr's own documentation points at, sold as two endpoints: algorithmic domain suggestion, and a status check in either a registry-level or a DNS-and-aftermarket scope.
| dotsweep | Fastly Domain Research API | |
|---|---|---|
| No key | YesNo key, no account, no signup. curl https://dotsweep.com/check/example.com works from a cold start. |
NoA Fastly account and a token sent as Fastly-Key. Their documentation adds that a user with the superuser role has to enable the product on the Products page before any call succeeds. |
| Admits doubt | YesA registry that is throttled, timed out or publishes no RDAP resolves to estimated, never available. A 429 from one registry does not become a confident answer about another one on the same host. |
YesTheir documentation says "If an upstream provider fails to respond within that timeframe, our API will return a best-effort value, typically unknown or undelegated". unknown is a real third state. undelegated is the same token a healthy DNS-level check returns, so a caller cannot tell that case from a confirmed absence of delegation. |
| Renewal + term | YesRegistration and renewal are separate figures, a first-year discount is flagged as a promotion, and an extension the registry sells only in multi-year terms is multiplied out so the leading number is what leaves your account. | NoThe status response documents aftermarket price fields, a currency, a price and a vendor, which is what a name is being resold for. The reference we read carries no registration price, no renewal price and no minimum term. |
| Eligibility | Partly38 extensions have had their registry policy read, 33 of them with an eligibility rule, each citing the registry's own page and the date it was read. An extension that is not in that set asserts nothing. | Not documentedTheir reference describes the status as a space-delimited string of statuses and does not enumerate them on the page we read. Domainr, whose vocabulary this product succeeds, publishes reserved, disallowed and claimed; we did not confirm that the successor carries them. |
| Free tier | Yes120 requests a minute, counted as requests rather than domains, and a batch endpoint carrying 10 domains each. No daily or monthly ceiling and no credit balance, so the rate is the whole quota. Edge cache hits are not counted at all. | Partly10,000 free requests a month on each of Status-Precise, Status-Estimated and Suggest, then $0.001 a request for Precise and $0.0005 for Estimated. The confident answer is priced at twice the estimate. |
| Assistant-ready | YesA remote MCP server at /mcp over the same handlers, addable as one URL with no account and no key, plus /llms.txt so an assistant that only fetches can still call it. |
Not documentedFastly publish an official MCP server for managing their platform. The pages we read do not show domain suggestion or domain status among its tools, and we did not test it. |
Where Fastly Domain Research API is the better choice
Written first because a comparison written by one of the two products is worth nothing if it cannot say this.
- Two availability scopes, priced separately: a registry-level Precise check and a cheaper DNS-and-aftermarket Estimated one. A caller chooses how much certainty to pay for, where we make that choice for you.
- Algorithmic domain suggestion across extensions and domain hacks, which is a different product from availability and one we do not offer at all.
- An ICANN-accredited registrar, a commercial support relationship and an invoice, none of which we have.
Where dotsweep is different
- No key, no account, no signup.
curl https://dotsweep.com/check/example.comworks from a cold start. - A registry that is throttled, timed out or publishes no RDAP resolves to
estimated, neveravailable. A 429 from one registry does not become a confident answer about another one on the same host. - Registration and renewal are separate figures, a first-year discount is flagged as a promotion, and an extension the registry sells only in multi-year terms is multiplied out so the leading number is what leaves your account.
- 38 extensions have had their registry policy read, 33 of them with an eligibility rule, each citing the registry's own page and the date it was read. An extension that is not in that set asserts nothing.
- A remote MCP server at
/mcpover the same handlers, addable as one URL with no account and no key, plus/llms.txtso an assistant that only fetches can still call it.
The unconfirmed-rather-than-available rule is the one worth explaining,
because it is the only wrong answer in this category that costs somebody money.
A registry that is rate-limiting you often phrases the refusal in words a parser
reads as "no such domain", so the naive result of a throttled lookup is a
confident "available" for a name somebody already owns. dotsweep asserts this in
its test suite in both engines, and marks the answer estimated
instead.
Is dotsweep better than Fastly Domain Research API?
Not at everything, and not for every job. Fastly Domain Research API does this better: Two availability scopes, priced separately: a registry-level Precise check and a cheaper DNS-and-aftermarket Estimated one. A caller chooses how much certainty to pay for, where we make that choice for you. dotsweep answers a different question: it needs no key or account, it returns the renewal price and the minimum term rather than the advertised first-year figure, it says who is allowed to register the extension, and it reports a throttled or unreachable registry as unconfirmed rather than available.
What Fastly Domain Research API documents
- https://www.fastly.com/products/domain-research-api, read 2026-08-01
- https://www.fastly.com/documentation/reference/api/domain-management/domain-research/, read 2026-08-01
- https://docs.fastly.com/products/domain-research-api, read 2026-08-01
- https://www.fastly.com/pricing, read 2026-08-01
- https://www.fastly.com/products/mcp-server, read 2026-07-31
How we know this, and what we did not do
Every claim about another product here was read from that vendor's own documentation, and each one is listed with the page and the date it was read. No figure comes from a review, a listicle or another comparison.
We did not benchmark any of these services. Nothing on this page reports a measurement, and where a row says "not documented" it means their documentation did not answer the question and we did not test it. It is not a quiet way of saying no. Getting that distinction wrong is how the rest of a comparison table stops being worth reading, and it is the same rule the rest of this site follows about extensions whose registry policy nobody has read.
If something here is out of date or wrong, the date next to it is there so you can tell which, and support@dotsweep.com reaches a person.
The one claim here that is measured rather than cited
Every row above comes from a vendor's own documentation. One thing does not, because no vendor documents it: what a checker concludes when the registry refuses to answer. That is testable, and the test is published with its method and a way to run it against your own — including why the obvious version of the experiment, waiting for a real rate limit and asking everyone at once, cannot work.
Check a name without signing up for anything
curl https://dotsweep.com/check/example.com