JoblumiBot

JoblumiBot is the fetcher of Joblumi, an app where people keep track of their own job applications. It fetches pages for a person's own job search, in three cases described below. Two happen only when the person clicks. The third, following an employer, checks a company's job list at most once a day.

It identifies itself with this user agent:

JoblumiBot/1.0 (+https://joblumi.com/bot)

When a person adds a link to a job ad to their board, JoblumiBot loads that page so the person doesn’t have to copy the text by hand.

  • It fetches exactly one URL: a job ad on a company’s careers page or website, or on a job site that isn’t on our block list.
  • It keeps only the text of the job ad, with the link and the date it was fetched. That text is visible only in the board of the person who added it.
  • Joblumi can check later if the ad is still online. This also happens only on that person’s click, at most once a day per job.

2. Company briefing

With a pass, a person can ask for a short briefing on the company behind a job ad, to prepare for an interview. The person enters the company’s website address, or picks the one suggested from the ad, and JoblumiBot reads that website.

  • It fetches at most 5 pages per click: the home page and up to 4 pages such as About us, Careers or Products.
  • It stays on the website of that one company and doesn’t follow links to other sites. If a page blocks it, it stops reading.
  • Each page is limited to 2 MB and 10 seconds. It reads only public HTML pages, no pages behind a login, no search or filter pages with parameters in the address, and no files.
  • It skips team and staff pages. The briefing names only your managing directors or other leaders, and the contact person in the job ad, with name and job title. It doesn’t name other employees and doesn’t use photos.
  • It deletes the page texts once the briefing is written. For that person only, it keeps the page addresses and titles and the checked statements about the company, each with a short quote from your page as its source. We delete them after 30 days.
  • The briefing is marked as written with AI support.

3. Following an employer

People with a Joblumi account can follow a company to hear about its new jobs. For this, JoblumiBot checks the company’s job listings on a schedule:

  • What it checks: first a public job feed or job board API that the company uses for its job ads (for example from its applicant tracking system), or job data in the page’s structured data. Only if there is none, the overview page of your job listings, at most 3 pages.
  • How often: at most once a day per company, however many people follow it. It checks only companies that at least one person follows.
  • What it keeps: job title, location, link to your job ad and the date it first saw the job. It doesn’t keep the text of your job ads or the names of contact persons, and it deletes a job 30 days after it is no longer listed. People who follow you get a short notice with a link to your own job ad.
  • What it doesn’t do: it doesn’t read single job ads on a schedule, doesn’t follow links, and doesn’t check job portals.

The rules on robots.txt, blocking and opting out below apply here too. If you block JoblumiBot, people who follow you see that you asked us not to check your jobs automatically.

What applies to all

  • JoblumiBot fetches pages when a person clicks, with one exception: “Follow an employer” (see 3). It doesn’t crawl your site.
  • It doesn’t log in, solve CAPTCHAs or use proxies, and it doesn’t pretend to be a normal browser.
  • It doesn’t fetch from LinkedIn, Indeed, StepStone, Glassdoor or XING, including their country sites and sister brands. People who find a job there paste the text themselves. The block list is checked after every redirect.
  • It doesn’t keep whole pages, screenshots, images or logos.
  • It doesn’t build a shared database. What it fetches is not pooled across users, not published, not searchable by others and not used to train AI models.
  • If a page answers with an error such as 401, 403 or 429, or shows a CAPTCHA or a bot check, JoblumiBot stops and doesn’t try again.
  • Our fetch logs keep only the domain and the time, and we delete them after 30 days.

Joblumi limits how many pages JoblumiBot fetches from one site, per minute and per day, across all users.

JoblumiBot follows robots.txt and TDM reservations

Before each fetch, JoblumiBot reads your robots.txt. If it blocks the path for all bots (User-agent: *) or for JoblumiBot, the page is not fetched.

It also respects a text and data mining reservation: the TDM Reservation Protocol (/.well-known/tdmrep.json, or the tdm-reservation header or meta tag) and noai in the robots meta tag or the X-Robots-Tag header. If one of them applies, JoblumiBot doesn’t read the page.

To block JoblumiBot from your whole site, add:

User-agent: JoblumiBot
Disallow: /

To block only part of your site, for example your careers section:

User-agent: JoblumiBot
Disallow: /careers/

How to opt out by email

You can also ask us to stop fetching your site. Write to hello@joblumi.com with the domain you want blocked. We put the domain and all its subdomains on our block list within 5 working days, and JoblumiBot won’t fetch any page on it after that, even when a person adds a link, asks for a company briefing or follows your company.

If your terms of use forbid automated access to your site, tell us the same way and we block your domain.

You don’t need to give a reason.

If you are named in a job ad or on a company website

Job ads often name a contact person, and company websites name managing directors. If that is you, you can object to us using your data or ask us to delete it: hello@joblumi.com.

Contact

Questions about JoblumiBot: hello@joblumi.com. Legal details are in the imprint, and how we handle personal data, including people named in job ads and on company websites, is in our privacy policy.