← Secure tools

Topic

Privacy-friendly search visibility

How PrivacyFileLab can be found in search while file checks stay in the browser, and what its robots, indexing, and analytics choices mean.

Public pages and private files

The pages this site expects crawlers to find are static HTML with a stable URL, a title, a heading, and links from other pages. The sitemap lists those URLs. A readable page does not need a script that identifies the reader. This topic, the tool pages, and the guides are that kind of HTML. Nothing here promises a position in search results.

The scanners read a file in a same-origin worker in the browser. They do not post the bytes to a PrivacyFileLab upload API, and they do not mint a public result URL from the file. Indexing /photo tells a crawler that the photo scanner exists. It does not publish the JPEG, PDF, or text someone selected in that tab.

Robots, indexing, and language URLs

robots.txt allows crawling of / and points at the sitemap on this host. That decision is about the HTML documents. Blocking every crawler would hide the documentation. It would leave a local scan unchanged, because the scan does not depend on the page being indexed. llms.txt repeats the titles and descriptions of the indexable pages and states the limits. It does not list the calculator that is kept out of the index. A line in that file still does not contain a file a visitor selected.

Most public URLs are indexable. /hvac-cost-comparison is the exception that stays on the site. The page is a retained heat-pump and furnace calculator, not a file tool. Its HTML asks crawlers not to index it, the response repeats that with the X-Robots-Tag header, and the sitemap omits the URL. The file-tool navigation and llms.txt do not list it. noindex asks crawlers to leave the URL out of the index. Anyone with the address can still open the page.

English pages have no /en/ prefix. Simplified Chinese pages live under /zh-CN/. There is no automatic redirect that uses only the browser language. Each public page sets a canonical URL for the language you are reading, plus hreflang alternates for en, zh-CN, and x-default. On this topic those paths are /topics/privacy-friendly-seo and /zh-CN/topics/privacy-friendly-seo. The tags tell a crawler which URL is which language copy. They do not name the person reading the page.

Analytics, referrers, and the request itself

The app responses set Content-Security-Policy with script-src limited to this origin, including inline scripts, and connect-src limited to this origin. They set Referrer-Policy: no-referrer, so a click from this site to another site does not send this page URL as the Referer. Permissions-Policy disables camera, microphone, geolocation, payment, and USB on these pages. The shipped HTML does not load an advertising script, a session-replay script, or a third-party analytics SDK.

The SHA-256 tool shows the digest on the page and does not send that digest to an analytics service. A full file hash is also outside what the analytics boundary of this project allows a measurement event to carry. The home page line "No tracking" means this app does not add a tracker. The network, the browser, and the host that serves the HTML can still see that a URL was requested. That is a different fact from a script that follows a person across sites.

Where the file tools fit

Search visibility and file privacy meet when a public page embeds or links to a file. A crawler that fetches an image URL receives that file, including metadata the format still carries. Before a JPEG, PNG, or WebP is placed on an indexable page, the photo scanner can list supported GPS, device, time, author, and XMP fields. Cleaning rewrites supported fields and scans again. It leaves faces, documents, and street signs in the pixels. The photo guide explains that limit.

A PDF linked from an indexable page can carry an author, earlier revisions, attachments, form answers, scripts, and external links. The PDF scanner reports the patterns it implements, up to 32 MB, and the cleaner refuses to rebuild many of those structures. DOCX, XLSX, and PPTX packages can carry identity, timestamps, and custom properties. The Office scanner is the local check for those packages.

The words on the page are what a search engine stores. A customer name, a credential, or a private URL in a heading or an image description is published when the page is indexed. The text redactor and the AI upload scanner review text locally before you paste it into a public page or into an AI product that will write one. The AI upload page does not call an AI API. A scan with no findings is not permission to publish the paragraph.

Do not try to identify who opened a scanner by joining ranking reports to file contents. This site does not receive the file. Readable titles, ordinary internal links, and the sitemap are the visibility tools these pages use.

A fictional example

Imagine a help page about sharing a JPEG. It needs a title, one H1, and a normal link to the photo scanner before it belongs in the sitemap. The sample image is a separate choice. Run that JPEG through the photo scanner first. If a supported GPS field is present, clean the supported metadata, scan the downloaded file again, and only then place that cleaned file at a public URL.

Write the image description without a real name, a home address, or a filename that identifies someone. The fictional page does not need a visitor-identification script to be linked from the home directory. The names in this example are not a case study, and the page does not claim that following these steps produces a ranking.

FAQ

Does the sitemap include files opened in a scanner?

No. The sitemap lists public page URLs such as /photo and the guides. Opening a file in a scanner does not add that file to the sitemap, and the scanner does not post the file to PrivacyFileLab.

Does noindex remove the cost calculator?

The page at /hvac-cost-comparison still opens. It is marked noindex, the X-Robots-Tag header repeats that, and the sitemap omits it, because it is a retained calculator rather than a file-privacy tool. The file-tool footer does not link to it. noindex is a request to crawlers, not a lock on the URL.

Does the lack of an analytics SDK make a scanned file anonymous?

This app does not load a third-party analytics SDK. A scan result is still not proof that a file is anonymous. Picture content, metadata this version does not list, and text you later publish can identify a person even when no analytics script ran.

On this site

Photo scannerThe photo scanner reads metadata containers in a JPEG, PNG, or WebP file. It looks for supported GPS, device, time, author, and XMP fields so you can review them before you share the file.PDF scannerThe PDF scanner inspects a PDF for metadata and structural signals such as document history, embedded files, forms, JavaScript, actions, and external links.Office scannerThe Office scanner opens DOCX, XLSX, and PPTX packages locally and looks for identity, timestamps, and custom properties in the package metadata.AI uploadThe AI upload scanner checks pasted text or a text file for common credentials, contact details, database URLs, IP addresses, and similar patterns before you paste that text into an AI product.Text redactorThe text redactor finds supported sensitive patterns, lets you choose which matches to replace, and checks the replacement text again.SHA-256The hash tool computes SHA-256 of the bytes you select so you can compare files. The hash is an identifier of those bytes, not a privacy verdict.Photo EXIF and GPS guideHow GPS, camera, and timestamp metadata can ride along with a photo, what the local scanner can detect or remove, and what it cannot see in the pixels.PDF metadata guideHow PDF metadata, revisions, attachments, forms, scripts, and links can expose more than the visible pages, and when a local cleaner refuses to rebuild the file.Guide to secrets before a pasteHow to review text for common secrets before pasting it into an AI tool, and why a clear pattern scan is not permission to upload the rest.