What this is
A workshop for documents, not a service that holds them.
MyPDFiles is 140 PDF tools that run inside your browser tab. You bring a file, the work happens on your own machine, you take the result away. Nothing is kept, because nothing is ever received.
Most sites in this category are thin interfaces over an upload endpoint. That design is cheaper to build and it scales to any file size, but it makes confidentiality a matter of policy — you are trusting a retention promise you cannot inspect. This site takes the harder route so that promise is unnecessary.
The one design decision
Processing locally is the whole point, and it costs us something.
A contract, a medical letter, a passport scan, an unsigned offer — these are the documents people actually run through PDF tools, and they are precisely the documents that should not be posted to a stranger’s server on the way to being rotated by ninety degrees.
So the work happens in your browser, in JavaScript and WebAssembly, on bytes read from your own disk. This is not a cheaper architecture. It is more difficult to build, it constrains what is possible, and it means we give up the analytics that come free when every file passes through your own server. We think that trade is obviously correct.
- What we gain
- Confidentiality that does not depend on trusting us, and a claim you can verify in your Network panel.
- What we give up
- Unlimited file sizes, server-speed OCR, and any conversion needing a native toolchain. Input is capped at 250 MB per file.
Absences on purpose
What this site deliberately does not do
Every item here is a feature someone has asked for. Each is missing for a reason we are willing to write down.
No accounts
Nothing here needs to know who you are, so there is no email address to collect and no password database to lose.
No cloud storage
No file history, no sync, no “recent documents” list. Closing the tab is the delete button.
No password cracking
We will not build brute-force or protection-bypass tooling. Removing a password you already know is a different thing from defeating one you do not.
No output watermarks
Your document comes back as your document. Branding somebody else’s file is not a growth strategy we are interested in.
No dark patterns
No decoy download buttons, no fake virus warnings, no countdown before your file is ready, no ads positioned to catch a misclick.
No invented certainty
Where a tool approximates — compression, accessibility checks, layout-preserving conversion — the tool page says exactly where it stops being reliable.
Who pays for this
How the hosting bill gets paid
Running this site costs money — hosting, a domain, bandwidth for the OCR engine — but it costs far less than a comparable service, because the expensive part of PDF processing is happening on your device rather than on a machine we rent by the hour.
The plan is display advertising. Ads are not enabled by default in this codebase; the slots reserve their space so that switching them on later cannot shove the page around. When they are enabled, an ad network is a third party with its own cookies, so they load only after you have accepted them in the consent banner. Decline and the tools work exactly the same — this is stated in full on the cookies page.
There is no paid tier, and no plan for one. A subscription would imply features held back behind it, and since your own browser does the processing there is no per-file cost for us to recover. What you can do here today is what everybody can do here.
- Price
- Free
- Paid tier
- None
- Sign-up
- Not required
Under the hood
How the thing is actually built
One registry, many pages
Every tool is a data record. Its page, metadata, structured data, search entry and hub listing are all derived from that one record, so there is no per-tool component drifting out of sync.
Engines loaded on demand
The homepage does not ship a PDF parser. Each tool imports only the engine it needs, which is why a page that rotates pages stays small and only the OCR tools pull down WebAssembly.
Validation before parsing
Files are checked by magic bytes rather than file extension, filenames are sanitised against path traversal, and size and expansion ratios are capped before any parser sees a byte.
The security page documents the headers, limits and guards in detail, with the real numbers imported from the code rather than retyped.
Reasonable questions
Who runs this site?
A small independent effort rather than a company with a floor of account managers. There is no venture funding behind it, no sales team and no enterprise tier. That is worth stating plainly because it shapes everything else: the tools are narrow and well-made rather than broad and shallow, and support is email rather than a live chat widget.
Why should I believe the no-upload claim?
Because you can inspect what happens without trusting us. Open your browser’s Network panel before running a tool: its code and assets arrive at your device, and consented content-free events may leave it, but your document is not uploaded. Before trying offline, first run the relevant tool with a non-sensitive sample so its engine and any required worker or language-model assets have loaded. Opening the page alone is not enough, and missing assets can prevent an offline run; this is not a guarantee of offline availability. A claim you can test is worth more than one you have to accept.
Is it really free, or is there a catch waiting?
It is free with no sign-up, no trial clock and no watermark on your output. The plan is to cover hosting with display advertising, which is disclosed up front rather than discovered later. There is no paid tier holding back features, because there is no per-file server cost to recover — your own device does the work.
Why are some obvious tools missing?
Some PDF work genuinely needs a heavy native toolchain — high-fidelity office-document rendering is the usual example. Doing that convincingly in a browser tab is not currently possible, and doing it badly while calling it done would be worse than leaving it out. Where a tool here has limits, the tool page says so on the page rather than in a footnote.
Will you add accounts, cloud storage or file history?
No. Each of those requires holding your documents, which would dismantle the one property that makes this site worth using. If you want your files synced across devices, a storage provider is the right tool; this is a workshop you bring a file to, not a place you keep files.