LogicDartBot

LogicDartBot is the program that Logic Dart uses to read web pages. If you found it in your server logs, this page explains who it is and how to stop it.

Who runs it

Logic Dart is owned and run by CSEAM Technology Corporation. To reach the people who run the bot, write to [email protected]. A person reads that mailbox.

How it identifies itself

Every request carries this user agent, and no other:

LogicDartBot/0.1.0 (+https://logicdart.com/bot; [email protected])

The version number changes as the bot changes. The bot does not use a browser user agent, does not rotate addresses, and does not use proxies. If your site refuses LogicDartBot, it stops. It does not try again under another name.

What it reads

Logic Dart helps a business understand how it appears online. When a customer adds their website, the bot reads that site's public pages: robots.txt, the sitemap, the main pages and the rest of the page list, up to a cap of 200 pages per visit. It also asks a small number of public sources for facts about the business's place, for example the US Census Bureau's address geocoder, under that source's own terms.

Only businesses and public pages. The bot does not look for private individuals and does not store the names of people it happens to find on a page.

What it never reads

Search engine result pages. The bot refuses to request them, from Google, Bing, DuckDuckGo, Yahoo, Yandex, Baidu, Brave Search, Ecosia, Startpage or any other search engine. It does not run searches.
Search pages inside other sites.
Addresses that are not on the public internet (private networks and internal names).
Anything your robots.txt disallows for LogicDartBot.

How it behaves

It reads your robots.txt before it reads anything else on your host, and follows it. It matches the name LogicDartBot, in any capitalisation. If robots.txt cannot be reached because of a server error, the bot reads nothing from that host on that visit.
It honours Crawl-delay. The wait between requests to one host is the larger of one second and the delay you state.
It makes one request at a time to a host, and never more than one a second.
If your server answers 429 or 503, it waits for the time in Retry-After; without one it waits 30 seconds, then 60, then 120, and then stops reading that host for the visit.
If your server answers 401, 403 or 451, it stops reading that host at once.
It fetches plain HTML. It does not run a browser in this version.
It writes a log line for every request: time, host, address, status, size, and the rule that allowed it.

What it keeps, and for whom

For a customer's own site, the bot keeps a compressed copy of each page it read, a fingerprint of the page, and the facts it extracted (such as the business name, address, phone number and hours). These belong to that customer and are used to show them what Logic Dart found and, later, to check that changes were made as intended.

For any page that is not the customer's own, the bot keeps the address, the time, the response status, a fingerprint, the facts taken from it, and a one-line note. It does not keep the page itself.

Every fact is stored with the address it came from and the time it was read, so the customer can see where it came from.

How to ask it to stop

Any one of these works:

Add a rule for LogicDartBot to your robots.txt, for example:
```
User-agent: LogicDartBot
Disallow: /
```
The bot sees it on its next visit.
Write to [email protected] with your site's name, and we will add it to a do-not-read list that the bot checks before robots.txt. We reply to say it is done.
Answer the bot with a 403. It stops reading that host at once.

If you think the bot misbehaved, send the log lines from your server and we will look at them.