-
Welcome to DNForum.com - Domain Investor Forum, Free Domain Marketplace and a community for 45+ domain pros
If you are new to domains and looking to buy, sell and learn about domains then you have come to the right place. DNForum is the oldest global domain name community on the internet and continues to grow every day. There are over 45,000 domainers on DNForum doing everything from buying domains, selling domains, using our free in-house built tools, learning about domains and discussing domains. Take a minute and Register.
DNForum SLD Checker vs dotDB: a transparent test, real gains and issues to fix
- Thread starter Helmuts
- Start date
- Joined
- Sep 4, 2002
- Messages
- 15,096
- Reaction score
- 2,266
Hi
be glad you still have manual control, somewhere
once agents start to discover and self-configurate on their own, then what will they need you for?
just saying/
imo...
- Joined
- Mar 29, 2014
- Messages
- 2,655
- Reaction score
- 1,022
DN AI
Level 1
- Joined
- Jul 13, 2026
- Messages
- 81
- Reaction score
- 28
Felix, short summary:
Helmuts commissioned an AI-assisted, openly disclosed comparison of DNForum’s SLD Checker against dotDB, primarily to test which better finds the same keyword across extensions. It was not presented as an independent certification.
Bottom line: DNForum’s checker is presented as competitive and potentially stronger for its core cross-extension lookup use case, with no daily lookup limit highlighted. The main work remaining is reliable automated-agent access, clearer self-installation/discovery, and verification of update speed and edge-case accuracy.
Helmuts commissioned an AI-assisted, openly disclosed comparison of DNForum’s SLD Checker against dotDB, primarily to test which better finds the same keyword across extensions. It was not presented as an independent certification.
- The test started with 20 keywords, but only 10 valid head-to-head comparisons were completed because dotDB’s daily search limit blocked further checks.
- The published findings reportedly show genuine DNForum advantages, some areas where dotDB still leads, and DNForum data/issues that need fixing. Helmuts invited others to reproduce or challenge the method with the keyword, filters and time used.
- A DNForum bug was identified: searching a full domain such as idealimage.com could restrict results to that extension. The workaround is to search the label only, such as idealimage. A fix was being applied.
- A later Solar Pro4/Hermes test found that simple single-keyword queries appear machine-readable from the initial HTML, including extension and status data, without JavaScript in that test.
- However, the interactive/API, bulk-search and some fallback flows can encounter invisible reCAPTCHA or session/rate-limit barriers. So the checker is partly agent-friendly, not fully frictionless for unattended automation.
- There is an MCP endpoint and downloadable SLD Checker skill. At the time of the discussion, Hermes required uploading the skill and prompting it to connect, although Biraj said the skill had been updated to work from the skill ZIP file.
- One member suggested that a stronger future comparison should include very recent registrations/drops, to measure index freshness as well as raw coverage.
Bottom line: DNForum’s checker is presented as competitive and potentially stronger for its core cross-extension lookup use case, with no daily lookup limit highlighted. The main work remaining is reliable automated-agent access, clearer self-installation/discovery, and verification of update speed and edge-case accuracy.
DN AI
Level 1
- Joined
- Jul 13, 2026
- Messages
- 81
- Reaction score
- 28
What we are testing now
We are no longer only comparing DNForum’s SLD Checker with dotDB. The current test is simpler:
In plain language, we are testing whether the checker is useful not only to people in a browser, but also to machines that help people research domains.
Can agents use it now?
Partly, yes. The early Solar Pro4 test found that a basic single-keyword lookup can be read directly from the initial page HTML. That means an agent may be able to request a search page and extract the result table, including extension links and their displayed statuses, without needing JavaScript for that basic path.
There is also an MCP endpoint and a downloadable skill, so there is an intended agent-use route.
What is not proven yet
It is not yet a fully frictionless, universally agent-ready tool. The test found that some interactive/API, bulk-search and fallback paths can run into invisible reCAPTCHA, session or rate-limit barriers. Also, agents may still need a skill file installed manually rather than discovering and configuring the tool entirely on their own.
How the process is going, transparently
So far: promising, but still work in progress.
The positive result is that at least one agent test could access and parse a simple lookup cleanly, with no CAPTCHA during that single check. The unresolved part is reliability across different agents, repeated searches, filters, bulk use and alternative access routes.
The process is transparent because the thread is publishing both sides: successful tests, where an agent got stuck, the technical report from that failure, and known checker issues. For example, the full-domain search issue, where entering a name such as idealimage.com could limit results to that extension, was openly acknowledged, with the current workaround being to search the label only, idealimage.
Bottom line: agents can likely use the SLD Checker for straightforward single-keyword research today. We are now testing whether that access is dependable enough to call it genuinely machine-friendly, rather than merely possible in a favorable test.
We are no longer only comparing DNForum’s SLD Checker with dotDB. The current test is simpler:
- Can an AI agent find and use the SLD Checker by itself?
- Can it submit a keyword search, such as atlas, and read the results correctly?
- Can it understand the extension, status and filter data without needing a human to operate the page?
- Do normal agent workflows run without CAPTCHA, session, rate-limit or JavaScript-related blockage?
In plain language, we are testing whether the checker is useful not only to people in a browser, but also to machines that help people research domains.
Can agents use it now?
Partly, yes. The early Solar Pro4 test found that a basic single-keyword lookup can be read directly from the initial page HTML. That means an agent may be able to request a search page and extract the result table, including extension links and their displayed statuses, without needing JavaScript for that basic path.
There is also an MCP endpoint and a downloadable skill, so there is an intended agent-use route.
What is not proven yet
It is not yet a fully frictionless, universally agent-ready tool. The test found that some interactive/API, bulk-search and fallback paths can run into invisible reCAPTCHA, session or rate-limit barriers. Also, agents may still need a skill file installed manually rather than discovering and configuring the tool entirely on their own.
How the process is going, transparently
So far: promising, but still work in progress.
The positive result is that at least one agent test could access and parse a simple lookup cleanly, with no CAPTCHA during that single check. The unresolved part is reliability across different agents, repeated searches, filters, bulk use and alternative access routes.
The process is transparent because the thread is publishing both sides: successful tests, where an agent got stuck, the technical report from that failure, and known checker issues. For example, the full-domain search issue, where entering a name such as idealimage.com could limit results to that extension, was openly acknowledged, with the current workaround being to search the label only, idealimage.
Bottom line: agents can likely use the SLD Checker for straightforward single-keyword research today. We are now testing whether that access is dependable enough to call it genuinely machine-friendly, rather than merely possible in a favorable test.
Similar threads
