Frequently asked questions

EditMyStuff is a privacy-first collection of free, browser-based file tools for students and professionals — a PDF editor, a PDF Form Filler, a Resume Builder, an ITR-1 tax calculator, an AI Error Debugger, AI Handwritten Notes, and image tools. File tools run in your browser, so your files are processed on your device; AI-powered features send only the text or images they need to an AI provider, and each tool says so.

The file tools are free, with no trial, no watermark and no paywall. AI-powered features come with free AI requests every day, shared by all the AI tools: 3 a day without an account, and 15 a day once you sign in. The day is India's, so they reset at midnight India time (IST). Without an account, your free AI requests are counted per network address, so people sharing one internet connection (the same Wi-Fi or office network) share them. Daily Talk is different: signed in, it gives one free AI attempt a day and never uses credits. After today's free requests, each AI request needs 1 credit. Signed in, you can buy credit packs on the Credits page: one-time purchases paid through Razorpay, and credits don't expire. Your own API key (added in AI settings) doesn't get around this: requests your key pays for don't use up your free requests. Once today's free requests are spent, a signed-in user still needs a credit for each request, even with their own key (the credit is refunded when your key pays); signed out, requests are refused until the reset.

No. Every tool opens without signing in; Daily Talk's AI feedback and saving or sharing System Design diagrams need it. Signing in with Google gives you 15 free AI requests a day instead of 3, Daily Talk's one AI attempt a day, keeps your Interview Prep progress and mock interviews in your account, and lets you save and share System Design diagrams. Interview Prep's Play tab is the exception: its progress is saved on this device only and doesn't sync to your account.

Live today: 20 tools — PDF Editor, Sign PDF, Image to PDF, Merge PDF, PDF Form Filler, Image Editor, Resume Builder, ITR-1 Calculator, AI Error Debugger, AI Answers as Handwritten Notes, Reverse Singing Challenge, GitHub Repo Doctor, Web QA Playground, Website → Design System, Generate System Design, Frontend Interview Prep, AWS Certification Prep, AI Presentation Maker, Daily Talk and Reel Studio. The full list is on the tools page. Reel Studio is in beta: it needs you to sign in and pays with bought credits. Several more — a Meeting Transcript Analyzer, Legal Notice Generator, Invoice Generator and others — are in development and marked "Coming soon" on the home page.

Yes. The tools are touch-friendly and mobile-responsive — they work on iPhone, Android and tablets. A few tasks (like precise PDF editing) are more comfortable on a larger screen, but nothing requires a desktop.

Any modern browser: Chrome, Firefox, Safari and Edge, on both desktop and mobile. It needs JavaScript enabled. Some optional features — like on-device AI field review in the PDF Form Filler — are only available in recent Chrome/Edge builds and are always optional.

We're actively building new tools — including a Meeting Transcript Analyzer, a Legal Notice Generator (India), a Salary Negotiation Coach, an Invoice Generator and more (see "Coming soon" on the home page). Follow the blog for launch announcements.

Use Send feedback — it's in the site footer and in every tool's header, and it doesn't need an account. You can attach up to 3 screenshots if it helps. We store your message, the page you were on, whether your screen is phone-, tablet- or desktop-sized, any screenshots, and your email address if you give one (signed in, it's filled in with your account's email), so we can review it and improve EditMyStuff. We read everything, and popular, well-explained requests tend to get built first — several current tools started as user suggestions.

No. File tools open, edit and export your file in your browser, on your device; the file is never uploaded or stored on a server. AI-powered features are the exception: they send only the text or images that feature needs to an AI provider, and each tool says so. For example, the PDF Form Filler sends a flat or scanned page's printed text and field labels to name its fields — never your answers — and Design System's optional AI Vision sends the screenshot you choose.

No. We don't sell data, and file tools never send us your files. Without an account, we don't know your name or email; when you use an AI feature, we store a coded form of your network address (not the address itself) with the tool and the time, to count your free AI uses. If you sign in, we keep your account (your name and email from Google); a record of each AI request (which tool, when, which AI provider and model answered and how long the text was, never what you wrote); your free-use counts, credit balance, credit history and any orders; Daily Talk's daily AI-attempt records; credit requests you send (besides you, only the site owner can read them, shown with your email); feedback you send while signed in; and what you choose to save: Interview Prep progress, your target company and role, the GitHub and LeetCode figures you connect, mock interviews and code-judge submissions, and System Design diagrams. Text you send to an AI feature goes to an AI provider to produce the answer. Google Analytics runs only if you allow it in Cookie settings; it sets a cookie that identifies your browser to Google, and it never sees your files. If your browser sends a Global Privacy Control signal, it stays off.

Your file stays in the browser tab only while it's open and is never uploaded. On a shared machine, finish your edit, download the result, and close the tab — that clears the document from memory. Some tools save what you type in this browser (Form Filler answers from fillable PDFs you download, saved signatures, resumes, notes), so clear the site's data before you leave a shared machine. If you signed in, sign out before you leave. Avoid leaving the tab open and unattended.

Two kinds. Essential storage is always on because the site needs it: your sign-in session if you use an account, your theme and your cookie choice, each tool's drafts and settings on this device, and, only after you connect GitHub in the GitHub Repo Doctor, a session cookie that lasts 8 hours. None of it is used to follow you around other sites. Google Analytics is optional: it sets its cookies only after you choose "Accept all" or turn Analytics on in Cookie settings, and it stays off whenever your browser sends a Global Privacy Control signal. The privacy policy lists what each tool sends and keeps.

Not yet. When it does, ads from Google AdSense will appear only on reading pages (the blog, this FAQ, the AWS service guides and the Interview Prep topic pages) and not inside the tools, where you open files or type. Google and its partners may use cookies on those pages to show ads and measure them; ads that aren't personalised still use cookies, for example to limit how often you see the same ad and to fight fraud. Your files and the text you type into the tools are never part of it.

Open Cookie settings at the bottom of any page to turn analytics on or off; turning it off stops Google Analytics and deletes its cookies. If you visit from the EEA, the UK or Switzerland, Google's consent tool on the reading pages asks before any advertising cookie is used; to change that answer later, open Cookie settings on one of those pages and choose Review ad choices. Anyone can turn off personalised ads in Google's ad settings, and opt out of some other companies' personalised ads at www.aboutads.info/choices.

EditMyStuff is built and run by Vikas Rajput, an individual based in Delhi, India. You can reach him at contact@editmystuff.in, or through Send feedback at the bottom of any page. The about page says more about why the site exists, and the contact page lists what to include for payment or data requests.

Signing in (creating an account) and buying credits require you to be 18 or older; "Do I need an account to use it?" above says what an account adds. Sign-in is with Google only, and Google shares your name, email address and profile picture with us.

There's no delete button yet. Email contact@editmystuff.in from your account's address, and your account and the data linked to it are deleted within 30 days; order and payment records needed for tax are kept. You can delete saved System Design diagrams and reset Interview Prep's cloud progress yourself, in those tools. The same address works if you want a copy of your account data or need it corrected. What the tools keep in your browser (drafts, saved signatures, resumes, notes) stays on that device until you clear the site's data there.

Partly. Once a tool has loaded and your file is open, editing happens locally, so a dropped connection doesn't stop you. There's no offline mode, though: reloading needs a connection, and some features download what they need the first time you use them — the PDF reader, export fonts, the signature pad, OCR and background removal — so use them once while you're online. AI features always need a connection.

The AI-powered features, which call an AI provider: the AI Presentation Maker, System Design, the Resume Builder's AI buttons, Mock Interview's optional AI feedback and AWS Prep feedback, the GitHub Repo Doctor, AI Handwritten Notes and the AI Error Debugger. The PDF Form Filler also needs a connection to name fields on flat or scanned PDFs. Web QA Playground and Website → Design System need one because they fetch the page you test. Everything else — PDF editing, filling fillable forms, image editing, building a resume, the tax calculator — keeps working if your connection drops after the page and the features you use have loaded. OCR and background removal download their engines the first time you use them, and Daily Talk downloads its speech model from this site once.

It depends on the tool. The PDF tools open PDFs; the PDF Form Filler also handles Word (.docx) and OpenDocument (.odt) forms; the Image Editor accepts JPG, PNG, WebP and AVIF, and Image to PDF accepts JPG, PNG and WebP. Exports include standard PDF, PNG images, and editable DOCX/ODT where relevant. Legacy binary .doc files need converting to .docx first.

There's no upload limit, because file tools don't upload your file — processing happens in your browser. The practical ceiling is your device's memory: everyday PDFs, forms and images are no problem, while very large files are limited by how much your browser can hold. A few tools set their own limits: the PDF Form Filler takes files up to 25 MB, Unlock PDF up to 100 MB, the Image Editor adds up to 50 images at a time and scales an image down to 4,000 pixels on its longest side, and Image to PDF scales images to 2,000 pixels.

Yes — you always keep your output. Tools export standard PDFs, and depending on the tool you can also save PNG images or editable DOCX/ODT. The AI Handwritten Notes tool exports PDF, PNG, or prints directly.

Some tools (like AI Handwritten Notes) auto-save your work to your browser's local storage so a refresh won't lose it. Because it's local, a draft lives on that specific device and browser until you clear your site data — export a PDF whenever you want a permanent, portable copy.

Yes. The file tools are free for personal and commercial use — generate resumes, fill forms, edit documents and images for work as much as you need; AI features come with free uses, then need credits. You own everything you create.

Not today. EditMyStuff is a web app that runs in any modern browser, so there's nothing to install. A public API and embedding aren't available yet — if you'd find them useful, let us know and it helps us prioritise.

It's optional — AI features work without a key, within your free uses. To use your own quota, open AI settings in any AI tool and add a key for Groq, Gemini, OpenAI or Anthropic; it's shared across tools so you only add it once. Requests your key pays for don't use up your free uses. Once your free uses for a tool are spent, a signed-in user still needs a credit for each request, even with their own key (the credit is refunded when your key pays); signed out, requests are refused. The key is kept in your browser (for the session unless you choose to remember it), sent with each AI request to our AI service only to call that provider, and never stored on our servers.

No. Like the other AI tools, System Design works without a key: its AI calls run through EditMyStuff's own server, which holds the API key, within your free uses. You can still add your own key in AI settings if you'd rather use your own quota.

Not unless you choose to save it. A generated architecture lives in your browser tab and is yours to edit, undo/redo and export. Saving a design (optional, for a shareable link) requires signing in, and only you can edit what you've saved.

No. Like System Design, the AI Presentation Maker works without a key: its AI calls run through EditMyStuff's own server, which holds the API key, within your free uses. You can still add your own key in AI settings if you'd rather use your own quota.

Just the content. The AI only ever writes structured slide content — titles, bullet points, table data, chart values, diagram nodes — never raw HTML, CSS or pixel positions. The template you pick controls every visual detail, so the result stays properly designed. See the AI Presentation Maker for the full FAQ on templates, editing and export.

Credits pay for AI requests beyond the free ones. Every AI tool shares the same free requests each day (3 without an account, 15 signed in, resetting at midnight India time), and after that each AI request uses 1 credit. The file tools never use credits, and neither does Daily Talk. Credits don't expire. If an AI request fails, or your own API key paid for it, the credit comes back to your balance.

Credit packs are one-time purchases on the Credits page, which you reach from your account menu once you're signed in: Starter, 15 credits for ₹49; Popular, 50 credits for ₹149; and Value, 150 credits for ₹399. If you need more than a pack, you can choose a custom amount from 151 to 7,500 credits, at ₹2.66 a credit. Each pack is a one-time purchase: nothing renews and there's nothing to cancel.

Payments go through Razorpay's checkout, which offers methods such as UPI and cards. Razorpay receives the order and the payment details you enter; EditMyStuff never sees your card or UPI details. We keep the order, its status, Razorpay's payment id and your credit history. Credits are added once Razorpay confirms the payment to our server.

Yes, for an unused pack: if you haven't used any credits from it, ask within 7 days of buying it by emailing contact@editmystuff.in from your account's address with the order or payment id. If the same order is paid twice, the second payment is refunded. Approved refunds are processed within 5–7 working days and go back to the payment method you used. The refund and cancellation policy has the details.

Your free requests come back at midnight India time. Signed in, you can also send a credit request from the notice that appears when they run out: say briefly what you're working on, and the site owner may grant you credits. Besides you, only the site owner can read your request.

Yes. When you redact text, it's rendered into an image layer with the black box permanently baked in and the underlying text deleted. Copy-paste and text extraction return nothing — it can't be reversed.

Pages you don't change are copied as they are, and the text, shapes and images you add are drawn on top, so they stay sharp. When you edit an existing line, its original text is removed from the file; if a line can't be taken out of the PDF's text, the editor tells you and saves that page as an image instead, so its text can't be selected or searched. Images you insert are scaled down to 1,600 pixels on their longest side if they're bigger.

A website QA audit is a pre-launch (or post-launch) check that a page is technically sound: its SEO metadata is present, headings and images are structured for accessibility, security headers are set, the viewport is configured for mobile, and internal links aren't broken. The Web QA Playground runs these checks automatically on any public URL and shows the evidence behind each finding. For a step-by-step walkthrough, see the pre-launch QA checklist.

It analyzes a public page's HTML and response headers and reports on SEO (title, meta description, canonical, Open Graph, headings, image alt, structured data), accessibility (missing alt text, unlabeled controls, heading order, document language, duplicate IDs, landmarks), resource performance (TTFB, HTML size, script/style/image counts and sizes), responsive configuration (viewport meta), security headers, and — optionally — a same-origin crawl for broken internal links. Every finding links to the exact evidence, and the 0–100 score is deterministic, not AI-generated.

Yes. It reports the page title and its length, the meta description, canonical URL, robots directives, the viewport tag, heading structure (H1/H2), images missing alt text, Open Graph and Twitter Card tags, JSON-LD structured data, and whether robots.txt and sitemap.xml are reachable. It's a technical/on-page SEO audit — it does not estimate keyword rankings or traffic.

Yes, for internal links. The optional crawler follows same-origin links from the page you test (up to a limit you choose, default 25) and flags any that return an error status, plus pages nothing links to. It's opt-in and polite — read-only, no re-visits, no aggressive crawling.

Yes. It reviews the response headers for HTTPS, HSTS, X-Content-Type-Options, clickjacking protection (X-Frame-Options or CSP frame-ancestors), Content-Security-Policy, Referrer-Policy and Permissions-Policy, and notes server version disclosure. This is a non-destructive, headers-only review — it never probes or attacks the site.

It runs safe, automated DOM checks: images without alt text, buttons and links without an accessible name, form controls without labels, skipped heading levels, a missing document language, duplicate IDs, positive tabindex and a missing main landmark. Automated checks catch common issues but cannot replace manual testing with a keyboard and a screen reader.

Not in this version. Field/lab metrics like Largest Contentful Paint and Cumulative Layout Shift require rendering the page in a real browser, which this tool does not do. Those metrics are shown honestly as "not available" rather than estimated. It does measure what a normal fetch can observe: server response time (TTFB), HTML size and resource composition.

No. It fetches and analyzes the page's HTML and headers; it does not render the page in a real browser or on real devices. So it reports responsive configuration (like the viewport meta tag) rather than screenshots at each device size, and it never claims device-verified results.

No. It fetches the public page through a server-side proxy that only reaches public hosts, forwards no cookies or credentials, and returns just the HTML and safe headers for your browser to analyze. We don't save the pages you test (our host's request logs record the address you enter), and no data is sent to any AI service — there is no AI in the tool.

Website → Design System reads a public site's HTML and CSS and extracts its observed visual language — a color palette, typography, spacing, border radii, shadows and any real CSS custom properties — then detects UI patterns and possible consistency issues and lets you tweak components in a live playground. Everything is backed by real CSS evidence. See how to extract a design system from a website for the full walkthrough.

Yes. It parses colors from the page's CSS and inline styles — hex, rgb/rgba and hsl, including gradient stops and CSS variables — normalizes them, de-duplicates, and shows each with its occurrence count and where it was used. Frequently used colors are labelled "frequently used," never given an invented semantic name like "primary."

Yes. It collects font families (including families declared through CSS variables), a sorted type scale of font sizes, font weights, line-heights and letter-spacing, and previews them in the site's primary font where available.

Yes. Real CSS custom properties found in the source (for example --brand-500 or --radius-md) are preserved exactly, with their names and values, and shown separately from values that were merely observed from usage. Generated export variables use an --ems-extracted- prefix so it's always clear which names came from the site and which the tool created.

Yes. It detects patterns like buttons, cards, inputs, links and badges from multiple signals — semantic element, class-name hints and the resolved CSS/structure — not tag names alone. Similar elements are clustered into variants, and each pattern carries a confidence and an evidence level (observed, suggested, inferred or confirmed). When evidence is weak it says so rather than guessing.

Yes — conservatively. It flags near-duplicate colors, rare spacing or radius outliers next to a dominant value, near-duplicate font sizes and shadows, and patterns that use conflicting values. It only reports where there is deterministic evidence, so intentional variation isn't flagged, and it never produces a subjective "consistency score."

Yes. You can copy or download the extracted values as CSS custom properties and as deterministic JSON. The export contains only design values and source metadata — never cookies, headers or private data.

Only partially. It inspects the HTML and CSS as served — static style blocks, inline styles and linked stylesheets. Styles injected at runtime by JavaScript may not be visible, and when that limits coverage the tool says so instead of pretending the extraction is complete.

No. It extracts observed values and evidence-based patterns from the public CSS — that is not the same as a team's official, documented design system. "Extracted" does not mean "official." Use it to understand and reuse a site's visual language, and treat the output as a well-evidenced starting point, not a canonical spec.