E1 · Run and save
20 minRun the audit on the home page and 2 pages of your site, and save it to raio-x.txt.
Done when: the file exists and has the Redirecionamentos, robots.txt, Sitemaps, and Páginas blocks.
Lesson 2 · 16 min
Read-only X-ray: the HTML the server delivers
A search engine doesn't see your site the way you see it in the browser: it reads the HTML the server delivers, almost always without running JavaScript. A 40-line script does that reading for you, without changing anything on the site, and in minutes it shows what no pretty dashboard will.
Start the lesson
In this lesson you will
- 1Download the inemaSEO method and run the audit
- 2Read redirects, robots.txt, sitemaps, and pages
- 3Find a page with no internal link in the HTML
- 4Turn the output into a list of findings
Goal: run auditoria.sh on your site and read each line of the output the way a search engine reads it.
01What the X-ray looks at
Read-only: the script uses curl and changes nothing. Tap each block.
Click each block in the diagram to see what it does.
It ran, saved, read. Nothing changes on the site; the X-ray is the picture before you touch anything.
02From download to list of findings
Tap each step.
1. Download
inemaSEO is public, under the MIT license. You need a terminal with bash and curl: Mac and Linux already have them; on Windows, use WSL or Git Bash.
Clone the repository and go into the folder.
Where it is
github.com/inematds/inemaseo-metodo, tools/ folder.
2. Run
Pass the main address and the page paths, separated by spaces. Save the output to a file so you can compare it later.
It takes less than a minute on a small site.
Tip
Choose a product or service page and a new page that you haven't seen on Google yet.
3. Read
Canonical empty or different from the address? Internal links near zero? Meta robots with noindex on an important page? Each one becomes a finding.
Don't fix anything yet: just take notes.
Typical finding
A catalog page that became "search only" in the browser and left hundreds of pages with no links in the HTML.
4. Note
For each finding, copy the line of output that shows it and the lesson in this course that covers it.
That list is your plan for the next few weeks.
03The X-ray commands
Copy and run in the terminal. Replace the address with your own site's.
Terminal · download the method
git clone https://github.com/inematds/inemaseo-metodo.git cd inemaseo-metodo
Terminal · run and save
bash tools/auditoria.sh https://www.seusite.com /pagina-1/ /pagina-2/ > raio-x.txt cat raio-x.txt
List of findings (template)
# X-ray of <yoursite.com> on <mm/dd/yyyy> | finding | output line | lesson | |---|---|---| | <empty canonical on /pagina-1/> | canonical: | 4 | | <page with no internal links> | links internos no HTML: 0 | 7 and course 2 |
Example from the repository
Excerpt of auditoria.sh run on 10/19/2026 (robots.txt shortened):
== Redirecionamentos http://www.inema.club/ -> 308 https://www.inema.club/ https://inema.club/ -> 308 https://www.inema.club/ https://www.inema.club/ -> 200 == robots.txt User-Agent: * Allow: / (…) == Sitemaps https://www.inema.club/sitemap.xml: 83 URLs, 60 anotações hreflang https://www.inema.club/courses-sitemap.xml: 100 URLs, 0 anotações hreflang == llms.txt: 200 == Páginas -- /cursos/ [200] title: Cursos de inteligência artificial, agentes e automação | INEMA.club canonical: https://www.inema.club/cursos/ hreflang: hrefLang="en" hrefLang="es" hrefLang="pt-BR" hrefLang="x-default" links internos no HTML: 334 -- /conhecimento/ [200] canonical: https://www.inema.club/conhecimento/ hreflang: links internos no HTML: 58
On 09/30/2026 (commit bc50d11), the /cursos/ page became "search only" and the 310 course cards lost the only internal link that came in the server's HTML. On screen, everything looked fine.
The fix (8229495) was a collapsed index inside <details> with all the links, with no visual change. The "links internos no HTML" line for /cursos/, above, is the snapshot after the fix.
- Looking only in the browser: what JavaScript builds may not exist for the bot.
- Auditing only the home page: the problem is usually on the catalog page or the new one.
- Fixing before writing it down: without the "before" snapshot, you can't tell if it got better.
04Practice: your X-ray
Keep raio-x.txt: lessons 4 to 8 compare against it.
E2 · Three findings
20 minRead the output and fill in the list of findings with at least 3 lines (if everything is fine, write down what you checked).
Done when: each finding has the output line copied and the lesson that covers it.
05Test what you learned
The explanation appears after you answer.
Question 1 of 4
Why does the X-ray use curl and not the browser?skill
Correct
The browser runs JavaScript and builds things the bot may not see. The X-ray shows the server's version, which is the basis for the search engine.Not that one
The browser runs JavaScript and builds things the bot may not see. The X-ray shows the server's version, which is the basis for the search engine.Review: What the X-ray looks atQuestion 2 of 4
The /cursos/ screen looked normal, but the course cards disappeared from the links. What showed the problem?
Correct
The browser search built the links with JavaScript; in the server's HTML they did not exist. The collapsed index brought the links back without changing the look.Not that one
The browser search built the links with JavaScript; in the server's HTML they did not exist. The collapsed index brought the links back without changing the look.Review: The X-ray commandsQuestion 3 of 4
Does the X-ray change anything on your site?
Correct
It's read-only. That's why you can run it on any public site, including a competitor's or a client's, before you change anything.Not that one
It's read-only. That's why you can run it on any public site, including a competitor's or a client's, before you change anything.Review: What the X-ray looks atQuestion 4 of 4
You found a problem in the X-ray. What's the next step in this lesson?
Correct
First the full picture, then the fix. The list of findings becomes the plan for the next lessons and the before-and-after comparison.Not that one
First the full picture, then the fix. The list of findings becomes the plan for the next lessons and the before-and-after comparison.Review: From download to list of findingsQuiz complete
You got {a} of {n} right. If you get them all right, the lesson is marked as complete.Pick an answer. The explanation appears after you try.
06Takeaways
Read-onlyauditoria.sh reads the HTML from the server with curl and changes nothing.
The line that catches the mostInternal links in the HTML: a page with no links barely exists to the search engine.
Outputraio-x.txt + list of findings with the line and the lesson.