408 free browser tools that never upload your files.
randomly.online | All tools | About | Privacy policy
Randomly.online is a website of 408 free tools for PDFs, images, text, spreadsheets, dates, calculations and developer work. Every one of them runs in your browser. There is no upload step and no account, because there is no server to upload to: the site is static files, and the work happens in JavaScript on the machine you are already sitting at.
That design choice is the reason the site exists. Most online converters take your file, put it on someone's server, do the job there and hand back a link. Yours sat on a stranger's disk for as long as their retention policy says. Here the file is read by the browser, transformed in memory, and written back out as a download. Close the tab and it is gone.
The trade is real and worth stating. Very large files are limited by your own RAM rather than a server's, and a few tools need a modern browser for WebAssembly. In exchange, nothing you open is ever transmitted.
Last verified: 4 September 2026. Counts in this document were measured against the deployed site on that date, not estimated.
| Tools | 408: 407 in the 10 categories, plus a standalone SIP calculator |
| Pages in the on-site search index | 431 |
| HTML files deployed | 495 |
| Files that leave your device | 0 |
| Accounts required | 0 |
| Hosting | Cloudflare Pages, static |
| Installable | Yes, as a PWA with a service worker |
| Licence | See LICENSE |
Each category has a hub page listing everything inside it.
| Category | Tools | Hub on the site | Guide in this repository |
|---|---|---|---|
| Date and time | 74 | all-date-time-tools | Date and time tools |
| Text | 67 | all-text-tools | Text tools |
| Calculators | 64 | all-calculator-tools | Calculators |
| Image | 47 | all-image-tools | Image tools |
| 42 | all-pdf-tools | PDF tools | |
| Developer | 33 | all-development-tools | Developer tools |
| Converters | 28 | all-converter-tools | Unit converters |
| Random generators | 25 | all-random-generators | Random generators |
| SEO | 20 | all-seo-tools | SEO tools |
| Excel | 7 | all-excel-tools | Excel tools |
Every category has a guide in the column on the right: all of its tools, grouped by the job they do, with a note on the awkward cases each set handles.
A further thirteen pages sit outside those ten categories, including Typing Speed Test, Transfer Files Between Android and iPhone, Direct Cast for watching a video with someone remotely, and a SIP and lumpsum calculator.
flowchart LR
A["You pick a file"] --> B["Browser reads it<br/>into memory"]
B --> C["JavaScript or WebAssembly<br/>does the work"]
C --> D["Result written to a Blob"]
D --> E["Download from<br/>blob: URL"]
style B fill:#1e293b,color:#fff
style C fill:#2563eb,color:#fff
A PDF merge is a good example. The page loads pdf-lib and pdf.js from the same origin, reads your files with the File API, builds the new document in memory, and hands the result to the browser as a blob: URL. Nothing in that chain has a network hop after the page itself loads.
Image tools use Canvas and WebAssembly the same way. The Excel tools run SheetJS in the tab. Timezone tools read a 3.4 MB static JSON table of cities from the site's own /data/ directory, so no geolocation lookup happens either.
You do not have to take the claim on trust. Open any tool, press F12, switch to the Network tab, then load a file and run it. You will see the page and its scripts load, and then nothing further. No request carries your file.
The stronger check is in the source. Across the 495 HTML files and 168 JavaScript files deployed on 4 September 2026, no code written for this site uses FormData or XMLHttpRequest, the two APIs a browser reaches for to post a file somewhere. A grep for them matches thirteen files. Nine are third-party libraries loading their own assets from this origin: pdf.js, pdf-lib, jsPDF, html2canvas, Tesseract OCR, SQLite WASM and MediaPipe. The other four are pages whose visible text names the two APIs, which is to say pages making this same argument.
One page on the site does talk to a backend, and it is named rather than hidden. The multiplayer Ludo game uses Firebase to match players. It handles no documents.
Analytics is Google Tag Manager at page level. It records that a page was viewed. It never receives a file, a filename or anything you type into a tool.
Yes, after the first visit. The site registers a service worker and ships a web app manifest, so Chrome, Edge and Brave will offer to install it, and Android will add it to the home screen like an app. Once installed, tools you have already opened keep working with no connection.
There is no separate download. The install is the website.
|
Dhruba Singha Roy, sole developer. QA and automation engineer at Cognizant, working in Playwright, Selenium, Java and CI/CD. Previously founded the AnimeType social app, and led a Google Developer Student Club of more than 1,000 members. Alumnus of Haldia Institute of Technology.
|
The site is built to be quoted rather than scraped for training, and it says so in machine-readable form.
/robots.txt allows every crawler and carries a Content Signals declaration of search=yes, ai-input=yes, ai-train=no. Search indexing is welcome. Fetching a page to answer someone's question with a citation is welcome. Using the text as training data is not.
/llms.txt describes every category in one file, and /llms-full.txt lists every indexable page with a one-line description, so an agent can find a specific tool without parsing a hub page.
Structured data is on the pages themselves, not only in the sitemap. Measured across the deployed HTML on 4 September 2026: 458 pages carry BreadcrumbList, 457 carry FAQPage, 437 carry WebApplication and 357 carry HowTo, holding 4,892 question and answer pairs between them. The useful thing to cite from a tool page is that explanatory text, not the tool interface, which needs a person and a file to do anything.
There is no API. A tool cannot be driven by fetching a URL with parameters, because the work happens in the browser after the page loads.
Open an issue on this repository with the tool URL and what you expected. Browser and version help, since several tools depend on WebAssembly and codec support that differs between them.