DEV Community: gpszxtzys

apis‑sendafun: Open‑Source Public‑API Directory on Cloudflare Workers

gpszxtzys — Tue, 25 Aug 2026 03:20:50 +0000

apis‑sendafun: Open‑Source Public‑API Directory on Cloudflare Workers

Looking for a quick way to browse and discover free public APIs?
I built apis‑sendafun, a lightweight open‑source API directory running entirely on Cloudflare Workers.

🔗 Live site: https://apis.sendafun.com
📦 GitHub Repository: https://github.com/yusheng‑zhang‑star/apis‑sendafun
📝 Release v0.1.0: https://github.com/yusheng‑zhang‑star/apis‑sendafun/releases/tag/v0.1.0

What it does

‑ Browse hundreds of free public APIs
‑ Instant search & filter by category
‑ One‑click copy API endpoints
‑ Real‑time health‑check shows working/broken status
‑ Multi‑language UI: English, Simplified Chinese, Traditional Chinese
‑ Simple built‑in visitor statistics

Tech stack

‑ Cloudflare Workers
‑ Lightweight frontend
‑ GitHub‑Actions for automatic dataset sync
‑ Ready‑to‑use wrangler.toml for self‑host deployment

You can fork this repo and deploy your own instance with your custom domain in minutes.

This is v0.1.0, more features will come in later releases.
Feedback, issues and pull‑requests are welcome.

Building apis.sendafun.com — Serverless Backend For SendaFun Ecosystem

gpszxtzys — Sat, 22 Aug 2026 14:59:45 +0000

Hi everyone,

I’d like to share the backend layer of my SendaFun side‑project: apis.sendafun.com, a dedicated serverless API gateway powering the whole multi‑subdomain creative tool ecosystem.

All services run on Cloudflare Workers stack, no traditional VMs or containers. The primary design goal is keeping heavy image‑processing logic inside user browsers whenever possible, while the API layer takes care of authentication, paywall validation, payment webhooks, rate‑limiting and cross‑subdomain isolation.

Why a standalone API subdomain?

Originally, I considered embedding backend logic inside each sub‑domain. Very quickly I hit several hard problems:

Avoid cross‑site permission leakage between independent sub‑sites (shop.sendafun.com, pattern.sendafun.com, smartimgkit.com and others).
Centralize paywall & subscription validation, so every frontend tool respects the same business rules.
Unify webhook handling for multiple payment providers.
Central rate‑limit, logging, CORS and security headers in one single place.

Splitting out apis.sendafun.com gives me clean separation: front‑end sites are static, all sensitive business logic lives on the dedicated API host.

Core responsibilities of apis.sendafun.com

Subscription & paywall validation: Verify monthly / yearly membership status. Enforce business boundaries: subscription only unlocks self‑generated templates, cannot access pre‑made shop marketplace assets. Non‑member restrictions (watermark output, PDF export locked).
Payment webhook receiver: Handle payment events from multiple payment gateways, update user entitlements without sharing user sessions across sub‑domains.
Rate limiting & abuse prevention: Prevent users from bypassing paywalls or over‑consuming backend resources.
Storage glue layer: Connect frontend tools to Cloudflare R2 storage, without exposing R2 credentials to browser clients.
Cross‑subdomain secure communication: Strict CORS configuration, only allow trusted sendafun‑family origins. Block unknown external origins.

Important architecture trade‑offs

Privacy‑first principle: User‑uploaded photos for pattern generation stay inside browser. The API does NOT receive user source images. It only receives small‑size computed parameters, thumbnails and metadata.
Strict sub‑domain isolation: Even though they share one API backend, each sub‑site remains independent. No shared browser session / cookie across sites.
Not a public open‑API: This gateway is built exclusively for SendaFun ecosystem front‑ends. Third‑party public API access is not supported for now.
Some related sub‑services are still under active development: shop marketplace endpoints, pattern generator backend helpers are being built out.

Tech stack

Cloudflare Workers as serverless runtime
Cloudflare R2 for object storage
Workers KV for lightweight user‑state and entitlement storage
Strict CORS / COOP / COEP security headers
Multiple payment‑provider webhook handling

This backend is tightly coupled with the multi‑subdomain tool ecosystem I wrote about in my previous post. The whole project is iterating step‑by‑step.

👉 Visit API gateway: https://apis.sendafun.com

I would appreciate any feedback or architecture suggestions.

Building SendaFun: A Multi‑subdomain Tool Ecosystem For DIY Creators

gpszxtzys — Sat, 22 Aug 2026 14:52:45 +0000

Hi everyone!

I want to share my ongoing side‑project: SendaFun, a collection of browser‑native creative tools built for DIY hobbyists.

The whole system runs on Cloudflare stack: Pages, Workers and R2 storage.
Most heavy‑duty logic runs inside end‑user browsers. User‑uploaded images will NOT be sent to backend servers whenever possible, to protect user privacy.

What we are building

This is a multi‑subdomain matrix, each site focuses on one creative workflow:

shop.sendafun.com — Digital pattern marketplace, sell ready‑made mosaic & cross‑stitch templates. Independent store, separated from subscription tools. Currently in development.
pattern.sendafun.com — Pattern generator for cross‑stitch, mosaic, crochet / knitting projects. Convert photos into craft patterns. Currently in development.
smartimgkit.com — Local‑run image processing toolkit, live and publicly available.
icons.sendafun.com — Icon asset library for creators, live and publicly available.

Business boundary (important design decision)

Monthly / yearly subscription only unlocks self‑made template generation. Subscribers cannot freely download pre‑made shop templates. Those are sold separately as one‑off purchases or bundle packs.
Non‑member users get watermarked output, and PDF export is locked behind subscription. Preview only for guests.
Shop marketplace is completely isolated from tool subscription. Two independent business modules.

Technical challenges I’m solving

Isolate permission & data between different sub‑domains, avoid cross‑site permission leakage.
Keep frontend bundle size reasonable while running image‑convert logic inside browser.
Balance free tier limitations and paid‑member benefits, prevent users from bypassing paywall.
Build internal cross‑linking between sub‑sites for organic traffic, without sharing user session across sites.

This project is still under active iteration.
If you like DIY craft, photo‑to‑pattern tools, you are welcome to try it out.

👉 Visit SendaFun：https://sendafun.com

I would love to hear your feedback and suggestions.

From Blurry to Crisp: 5 Free AI Image Enhancement Tools That Actually Work in 2026

gpszxtzys — Mon, 29 Jun 2026 04:41:28 +0000

Adobe just acquired Topaz Labs for a rumored nine-figure sum. The message is clear: AI-powered image enhancement — not just generation — is where the industry is heading.

But Topaz Gigapixel costs $199. Adobe Creative Cloud runs $60+/month. What if you just want to fix a few blurry photos without breaking the bank?

Here are 5 free AI image enhancement tools that deliver real results in 2026. No subscriptions, no watermarks, no catch.

1. SmartImgKit Image Upscaler — Best free upscaler

What it does: Enlarges images by 2x, 3x, or 4x using smart interpolation + sharpening. Handles JPG, PNG, and WebP.

Best for: Printing phone photos, restoring old family pictures, making wallpapers from low-res game screenshots.

How to use it:

Go to smartimgkit.com/tools/image-upscaler
Upload your image (drag & drop or click)
Pick your scale (2x/3x/4x)
Download the result — all processing happens in your browser, nothing is uploaded

The quality is surprisingly good for a free tool. You won't get Topaz Gigapixel's AI model depth, but for most everyday use cases — printing 4×6 photos, posting on social media, prepping images for a website — it's more than adequate.

2. SmartImgKit AI Image Enhancer — One-click photo fix

What it does: Automatically adjusts brightness, contrast, saturation, and sharpness. Think of it as "auto-tune for photos."

Best for: Dark vacation shots that need quick rescue, flat product photos, batch-processing dozens of images.

The "Auto Enhance" button does a solid job with most photos. You can also tweak individual sliders if you want fine control. All 100% browser-based, zero server uploads.

Try it: smartimgkit.com/tools/image-enhancer

3. SmartImgKit Background Remover — AI-powered, runs locally

What it does: Detects the main subject and removes the background in seconds. Powered by WebAssembly + ONNX, running the AI model directly in your browser.

Best for: E-commerce product photos, profile pictures, presentation slides, any scenario where privacy matters (since nothing leaves your machine).

The key differentiator here is privacy. Most online background removers upload your image to their servers. SmartImgKit keeps everything local.

Try it: smartimgkit.com/tools/background-remover

4. The old photo restoration combo

Restoring old photos usually requires three steps: scan → enhance → colorize. Here's how to do the first two for free:

Step 1: Scan the photo at 300-600 DPI
Step 2: Use Image Upscaler at 2x to get more resolution
Step 3: Hit AI Image Enhancer to boost clarity and contrast
Step 4: Apply a warm tone filter from Image Filters if you want a vintage look

For the hardcore crowd: GPEN (an open-source face restoration model on GitHub) remains the go-to for facial detail reconstruction in 2026.

5. The full toolkit — 28 free tools in one place

All the tools above are part of SmartImgKit, which currently offers 28 free image utilities:

Image Compressor — Reduce file size without noticeable quality loss
Image Converter — JPG ↔ PNG ↔ WebP ↔ GIF ↔ BMP ↔ TIFF
HEIC Converter — iPhone photos to JPG/PNG
PDF to Image — Convert PDF pages to images
Watermark Tool — Text + image watermarks
15+ Image Filters — Grayscale, sepia, vintage, vibrant…
Meme Generator — Use templates or your own images
GIF Editor — Combine frames, adjust timing
OCR Text Extraction — 100+ languages
Bulk Image Processor — Batch resize/convert/watermark
Collage Maker, Circle Crop, SVG to PNG, EXIF Remover, Image Splitter…

Every tool is free, no registration, no server uploads.

👉 smartimgkit.com

Why browser-based AI image processing matters

There's a broader trend worth noting: AI image processing is increasingly happening in the browser, not the cloud.

SmartImgKit's background remover runs WebAssembly + ONNX locally. The upscaler processes everything client-side. This means:

Zero server costs — no usage limits, no paywalls
Privacy by design — your images never leave your computer
No latency — instant results, no waiting for cloud round-trips

This is the same direction the industry is heading. Adobe's acquisition of Topaz Labs validates "enhancement AI" as the next frontier. Meitu's 2026 Image Festival launched Picchi, a portrait editing Agent that learns your style.

The difference? Most of those are paid, walled gardens. SmartImgKit is free, open-concept, and runs entirely in your browser.

SmartImgKit is an independent open-source project. Not affiliated with Adobe, Topaz Labs, or Meitu. All tools run client-side in the browser — no images are uploaded to any server.

Why I Built a Free Image Tools Site (And How Free Can Actually Work)

gpszxtzys — Sun, 28 Jun 2026 04:14:56 +0000

I saw the "I built 184 free browser tools" post on HN last week — 88 points, solid discussion. A lot of comments were asking the same questions: "How do you make money?" "Why would you do this for free?" "What's the catch?"

I built SmartImgKit, a free online image tools site with 20+ tools — compressor, converter, resizer, background remover, watermark tool, meme generator, GIF editor, and more. All browser-based, no uploads, no signup.

I've been thinking about the same questions. Here's where I landed.

The gap in the market

The image editing market is split between heavy desktop software (Photoshop, GIMP, Affinity) and SaaS tools with monthly subscriptions. Neither works well for the person who needs to crop one image, remove a background from another, and convert a HEIC file to JPEG in the same hour.

Desktop tools require installation and updates. SaaS tools require signup and often upload your files to their servers — a genuine privacy concern for sensitive images.

Browser-based tools handle all of that. WebAssembly makes it possible to run complex processing — background removal with ONNX models, HEIC decoding, image upscaling with smart interpolation — entirely in the browser. Your files never leave your machine.

How free can work

A few sustainability models I've seen work in this space:

The Pro plan model. Free tier handles 80% of use cases. Paid tier adds batch processing, higher resolution output, or API access. This is what most online image tools do, and it works.

The enterprise model. Free for individuals, paid for teams with SSO, audit logs, and compliance features. Figma made this famous.

The adjacent-product model. Free tools drive awareness for a paid product in a related space. Not applicable to every project, but powerful when it fits.

I'm still figuring out the right model for SmartImgKit. But one thing I'm sure about: if the tools are genuinely useful and respect user privacy (no uploads, no unnecessary tracking), people will use them. Monetization can come later.

What I'd love to hear

What's the line for you between "this should be free" and "I'd pay for this"?

Is it volume (more than X images per month)? Features (batch processing, API access, higher resolution)? Or something less obvious like support, SLA, or white-label options?

Drop your thoughts in the comments. I'm genuinely curious where the community draws that line.

I switched 23 sites from JPEG to WebP/AVIF last month — here's what I learned

gpszxtzys — Sun, 28 Jun 2026 03:19:29 +0000

I spent last month migrating 23 client sites from JPEG/PNG to WebP and AVIF. Here's what I wish someone told me before I started.

AVIF vs WebP: the real numbers

AVIF is about 30% smaller than WebP at the same quality level. But Safari support is still patchy — if your traffic is 40%+ iOS, you need <picture> tags with WebP fallback. No way around it.

The biggest win wasn't the format

The single biggest reduction came from capping max image width at 1200px and setting quality to 80. One site went from 9.4MB to 318KB per page — a 97% reduction — just from those two settings plus lazy loading.

The format switch was the cherry on top, not the cake.

Tools I used daily

SmartImgKit — quick batch conversions in the browser. No uploads, no signup, drag and drop. Handles the 80% case where you don't need a CLI pipeline. Supports JPG, PNG, WebP, AVIF, GIF, BMP, TIFF.

ImageMagick — server-side batch jobs for when you need automation.

Squoosh — one-off fine-tuning with visual comparison.

Sharp (Node.js) — build pipeline integration.

The HEIC surprise

Every iPhone user's photos are HEIC. Most web tools crash on them. You need a converter that handles them before the pipeline — SmartImgKit's HEIC converter works locally in-browser, no uploads.

The 80/20 rule

Format + max width + lazy loading = 80% of the gain. Everything else is diminishing returns. Don't over-engineer it.