Cycling through all six. Tap any point to stop.

Measured on six things

Free PDF editor and file converter that runs in your browser without uploading your files.

Security

The no-upload architecture is a genuine security win — files never touch a server, so there's no retention risk. But the HTTP probe found 2/12 checklist points: no HTTPS enforcement, no HSTS, no CSP, no frame protection. SPF and DMARC are present.

Screenshots

1 / 3

EZPZ File

About this app

EZPZ File is a set of free file tools that run entirely in your browser. Your files are processed on your own device and are never uploaded to a server, so contracts, resumes and other private documents stay with you. What you can do: - Edit PDFs: change the existing text in place with the original font, add text, images, signatures and highlights, and reorder or delete pages - Merge, split, rotate, compress, password-protect and unlock PDFs - Convert PDF to Word, Excel, JPG and PNG, and images to PDF - Open Korean HWP and HWPX documents and convert them to PDF, or convert PDF back to HWP/HWPX - Convert HEIC and WebP images, resize and compress images, remove backgrounds and strip photo metadata - Extract ZIP files, open EML emails and convert XLSX to CSV No sign-up, no watermarks and no usage limits. Once the page has loaded, the tools keep working offline. Available in 20 languages.

Key features

1Client-side WebAssembly processing: files never leave the device, no upload, no server retention
2Broad tool coverage: PDF editing, merge/split/compress, PDF↔Word/Excel/JPG/PNG, HWP/HWPX conversion, image tools, archive extraction
3No signup, no watermarks, no usage limits, works offline after page load, available in 20 languages
4MCP integration for AI agents: one-line install (claude mcp add ezpzfile -- npx -y ezpzfile-mcp)

Target customer profile

Drawn from the product itself, not from a survey.

Scout Review

Written by AI from measured evidence, scored out of 100.

EZPZ File is a browser-based file tool suite with a genuinely strong architectural bet: everything runs client-side via WebAssembly, so your files never touch a server. That's a real differentiator in a category full of upload-and-pray converters. The tool list is broad — PDF editing, conversion, HWP/HWPX support, image tools, archive extraction — and the no-signup, no-limits model is exactly right for a utility. Usability is the standout: tools are organized by input format, the MCP one-liner is a smart hook for AI agents, and the friction-to-value ratio is near zero. But the product has real gaps. Mobile Lighthouse performance is 69/100 with an 11.3s LCP, which is a bad first impression. The security headers are nearly absent (2/12 checklist points) despite the privacy-first pitch. The design is functional but anonymous — no screenshots, no logo, no visual demonstration of the core claim. Accessibility is strong at 95/100 with minor fixable issues. Growth potential is real but the moat is thin. This is a competent, genuinely useful tool that needs to tighten its operational fundamentals and sharpen its visual storytelling.

Measure by measure

Usability

Zero-friction model: no signup, no watermarks, no limits, tools grouped by 'what file do you have'. MCP one-liner for AI agents and a how-to column add real depth. Navigation is genuinely thoughtful.

Design

Functional, text-dense directory layout organized by input format. No hero imagery, no product screenshots, no logo mark — reads more like a documentation index than a product landing page. Clean but visually anonymous.

Speed

Desktop Lighthouse 86/100 is solid, but mobile 69/100 with an 11.3s LCP is a real problem — that's four times past the 'good' threshold. CLS 0 and TBT 105ms show the layout is stable and JS isn't the bottleneck.

Security

Accessibility

Lighthouse 95/100 with only three failing audits: unnamed buttons, label mismatches, and table headers missing on a large table. 20-language support is a real accessibility win. No WCAG statement published.

Growth

Real problem, specific wedge: no-upload privacy plus HWP/HWPX support for the Korean market plus MCP for AI agents. Differentiated and aimed at a category with headroom, but the wedge is copyable and the moat is thin.

Improvement areas

Security headers: only 2/12 checklist points — no HTTPS enforcement, HSTS, CSP, frame protection, or Referrer-Policy
Mobile performance: 69/100 Lighthouse with 11.3s LCP — four times past the 'good' threshold
Accessibility: three failing audits (unnamed buttons, label mismatches, missing table headers) despite 95/100 score
Design: no screenshots, no logo, no visual demonstration of the core no-upload claim on the homepage

Design & Usability

EZPZ File's design is functional but anonymous. The homepage is a dense directory of roughly 40 tools grouped by input format — 'what file do you have?' — which is a genuinely smart navigation model that beats the usual grid of icons. But there's no logo mark, no hero imagery, no product screenshots, and the live counters ('387 files converted', 'editor used 1,548 times') read as small numbers rather than social proof. Usability is where the product shines: no signup, no watermarks, no usage limits, and a one-line MCP install for AI agents. The how-to column adds real depth. The friction-to-value ratio is close to zero, which is exactly right for a utility tool. The gap is that the design doesn't communicate the product's strongest asset — the no-upload architecture — visually. It's all text, no demonstration.

Speed & Security

Speed is a split verdict. Desktop Lighthouse performance is 86/100 with CLS 0 and TBT 105ms — the layout is stable and JavaScript isn't the bottleneck. But mobile performance drops to 69/100 with an 11.3s LCP, which is four times past the 'good' threshold. For a tool that runs entirely in the browser, mobile load time is the first impression, and 11.3 seconds is a bad one. Security is architecturally strong but operationally weak. The core claim — files never leave the device, WebAssembly runs in-browser, no server copy exists — is a genuine privacy win that most competitors can't match. But the HTTP probe found only 2/12 checklist points: no HTTPS enforcement, no HSTS, no CSP, no frame protection, no X-Content-Type-Options, no Referrer-Policy, no Permissions-Policy. SPF and DMARC are present. The architecture is paranoid; the headers are naked. For a product whose entire pitch is privacy, that's a credibility gap.

Accessibility & Growth

Accessibility scores 95/100 on Lighthouse with only three failing audits: buttons without accessible names, visible text labels not matching accessible names, and a large table missing <td> headers. These are fixable in an afternoon. The 20-language support is a real accessibility win that most indie tools skip. No WCAG conformance statement is published, which is a missed opportunity for a product targeting privacy-conscious users. Growth potential is solid but not spectacular. The problem is real and universal: converting files without uploading them to a stranger's server. The differentiators are concrete — no upload, no signup, no limits, HWP/HWPX support for the Korean market, and MCP integration for AI agents. The Korean document angle is genuinely underserved; Hancom Office is the only real player and it's not browser-based. The MCP hook is clever positioning for the AI-agent era. But the wedge is copyable: any competitor can add a WebAssembly build. The moat is thin, and the traction is minimal (387 files converted). The idea has room; the execution needs to earn it.

Conclusion

EZPZ File is the kind of tool you bookmark and forget you have until you need it — and then you're glad it exists. The no-upload architecture is the right bet, the tool coverage is impressive for a free product, and the Korean HWP support is a genuine niche advantage. But 'free and private' isn't enough on its own. The 11.3s mobile LCP will lose users before they see the value. The missing security headers undercut the privacy pitch. The anonymous design fails to demonstrate the one thing that makes this product different. None of these are hard problems. Enforce HTTPS, add HSTS and CSP, compress the mobile payload, name your buttons, and put a screenshot of the editor on the homepage. Do that and EZPZ File goes from 'useful bookmark' to 'the tool I recommend when someone asks how to edit a PDF without uploading it.' The foundation is there. The polish isn't. Yet.

Suggestions for improvement

Enforce HTTPS and add HSTS, CSP, X-Content-Type-Options, and Referrer-Policy headers — 2/12 is not acceptable for a privacy-first product
Fix the 11.3s mobile LCP: compress images, defer non-critical JS, and consider a lighter mobile layout for the tool directory
Add accessible names to all buttons and table headers — three Lighthouse failures are fixable in an afternoon
Put a screenshot or short demo of the PDF editor on the homepage — the no-upload claim needs visual proof, not just text

How we score

Alternatives comparison

Named competitors, point by point. Nobody paid to appear here or to be left out.

	EZPZ File
Processing location	Entirely client-side via WebAssembly; files never uploaded	Not specified in provided data
Signup required	No — no signup, no watermarks, no usage limits	Not specified in provided data
Korean HWP/HWPX support	Yes — HWP/HWPX to PDF and back, HWPX to text, HWP to HWPX	Not specified in provided data
MCP / AI agent integration	Yes — one-line MCP install for Claude and Cursor	Not specified in provided data
Offline capability	Yes — tools keep working offline once the page has loaded	Not specified in provided data
Language support	20 languages	Not specified in provided data