DEV Community: deepesh verma

How to Upload More Images PDFs, Docs to AI Tools (Without Hitting File Limits)

deepesh verma — Fri, 03 Apr 2026 15:27:32 +0000

If you’ve worked with AI tools like ChatGPT, Claude, or Gemini, you’ve probably run into this:

“You can only upload X files at a time.”

It sounds like a small limitation — until you actually need to:

Upload multiple screenshots
Share several PDFs
Provide full context for analysis Chrome - https://chromewebstore.google.com/detail/ai-upload-booster/hgimhlchljgiodcgjegcbfplbpfkhbgi?authuser=0&hl=en Firefox - https://addons.mozilla.org/en-US/firefox/addon/ai-upload-booster/

And suddenly, you’re stuck splitting your workflow into multiple prompts.

🚨 The Real Problem

Modern AI tools are powerful — but they’re constrained by file upload limits.

This creates real friction:

You can’t upload all your screenshots at once
You have to split PDFs across multiple messages
Context gets broken across prompts
The AI loses continuity in analysis

So instead of one clean interaction, you end up doing:

Upload a few files
Ask a partial question
Upload more files
Re-explain context
Repeat

It’s inefficient — and honestly, frustrating.

💡 A Smarter Way to Handle File Limits

Instead of uploading files one by one…

👉 What if you could combine multiple files into a single upload?

That means:

Multiple images → one image
Multiple PDFs → one document

So from the AI’s perspective:

You’re still uploading “one file” — but it contains much more information.

🖼️ Upload Multiple Images in One Go

One of the biggest pain points is image uploads.

Let’s say you want to:

Compare UI screenshots
Analyze charts or graphs
Provide multiple visual references

Normally:

You hit upload limits quickly
Or you send images across multiple prompts

Better approach:

Combine all images into a single high-quality image.

Now the AI:

Sees everything together
Maintains full visual context
Gives better, more consistent responses

📄 Upload Multiple PDFs Together

Same issue with documents.

If you’re working with:

Research papers
Reports
Notes

You often need all of them together for meaningful analysis.

Instead of:

Uploading 1–2 files per prompt

You can:

Merge them into a single document
Upload once
Ask one complete question

🧠 Why This Improves AI Responses

AI models perform better when they have:

Complete context
Fewer interruptions
Structured input

By combining files:

You reduce prompt fragmentation
You improve response quality
You save time

⚡ The Tool I Built for This

To solve this, I built a browser extension:

AI Upload Booster

It lets you:

Merge multiple images into one (for visual analysis)
Combine PDFs and documents into a single file
Prepare uploads directly in your browser

Everything happens locally — no uploads to external servers.

🛠️ How It Works

Select images or documents
Add multiple files
Merge instantly
Upload the combined file

That’s it.

🔒 Privacy Matters

100% local processing
No tracking
No file storage
No external APIs

Your data stays in your browser.

💭 Final Thoughts

File upload limits aren’t going away anytime soon.

But your workflow doesn’t have to suffer because of them.

By simply changing how you upload files:

You can include more context
Get better AI responses
Work faster with fewer prompts

🚀 Try It Yourself

If you regularly work with:

Screenshots
PDFs
Multi-file inputs

You’ll notice the difference immediately.

📢 Feedback

If you try this approach (or the extension), I’d love to know:

Does it improve your workflow?
What features would make it better?

That’s how tools like this evolve.

I Built a JSON Viewer with Table View, JWT Decoder, and Image Preview (Because Existing Ones Weren’t Enough)

deepesh verma — Thu, 02 Apr 2026 20:32:52 +0000

If you work with APIs daily, you already know this:

Most JSON viewers are… basic.

They pretty-print JSON.
They collapse/expand nodes.
And that’s about it.

But modern API responses aren’t simple anymore — they’re large, nested, and full of data that needs interpretation, not just formatting.

So I built JSON Vision Pro.

🚨 The Problem

While working with APIs, I kept running into the same issues:

Large JSON arrays were hard to read
JWT tokens required external tools to decode
Image URLs had to be opened manually
Comparing API responses was painful

Every time, the workflow looked like this:

Open JSON
Copy something
Go to another tool
Come back
Repeat

It breaks focus. It wastes time.

💡 The Idea

What if a JSON viewer could understand the data, not just display it?

Instead of treating everything as text:

Arrays become tables
Tokens become readable
URLs become previews

That’s the core idea behind JSON Vision Pro.

🚀 What JSON Vision Pro Does

📊 Table View for Arrays

JSON arrays are automatically convertible into clean table views.

No more scanning through nested objects — just read it like structured data.

🔑 Built-in JWT Decoder

Paste or detect a JWT → instantly see:

Header
Payload
Expiration

No more jumping to external websites.

🎨 Smart Visual Previews

The extension automatically detects and renders:

🖼️ Image URLs → inline thumbnails
🎨 Color codes → color chips
⏱️ Unix timestamps → readable dates

This alone removes a lot of manual effort.

⚖️ Response Comparison

Compare two API responses side-by-side:

See what changed
Highlight differences instantly

Super useful for debugging and QA.

🔍 Search & JSONPath

Search across entire JSON
Use JSONPath queries like:

$.users[*].name

Works well even with large payloads.

🕒 Time Machine (History)

Quickly switch between recent API responses and compare them.

🧠 Why This Matters

Most tools treat JSON as text.

But developers don’t think in raw JSON — they think in:

Data structures
Patterns
Differences

The goal was simple:

Reduce friction between “seeing data” and “understanding data”

⚡ How to Use

Install the extension
Open any API endpoint returning JSON
The viewer activates automatically

Works with:

REST APIs
Local JSON files
Any application/json response

🔒 Privacy First

100% local processing
No tracking
No data leaves your browser

💭 Final Thoughts

This started as a small frustration.

But once you start working with:

Table views instead of raw arrays
Inline previews instead of guessing
Built-in tools instead of switching tabs

…it’s hard to go back.

🚀 Try It Out

If you work with APIs regularly, I’d love your feedback.

JSON Vision Pro is available for Edge and Firefox.

📢 Feedback

If you try it:

Let me know what’s missing
Suggest features
Or just tell me how your workflow feels now

That’s how this gets better.

I Built JSON Vision Pro a JSON Viewer with Table View, JWT Decoder, and Image Preview (Because Existing Ones Suck)

deepesh verma — Thu, 02 Apr 2026 20:27:22 +0000

Created a tool for web developers

deepesh verma — Fri, 27 Mar 2026 14:16:54 +0000

https://addons.mozilla.org/en-US/firefox/addon/json-vision-pro/

Turns ugly raw JSON into a beautiful, interactive viewer with special tools for developers.

Core Features

Auto JSON Formatter - Beautiful color-coded tree view

Dark Professional Theme - Easy on the eyes

Collapse/Expand Nodes - Navigate complex structures easily

Copy JSON Paths - One-click path copying

Color Previews - See color chips for hex codes

Image Thumbnails - Preview images inline

Timestamp Converter - Unix timestamps → readable dates

Instant Text Search - Filter data in real-time

JSONPath Queries - Advanced search with $.users[*].email syntax

Table View - Convert arrays to sortable spreadsheets

Column Sorting - Click headers to sort

CSV Export - Download as Excel-compatible files

JWT Decoder - Decode tokens with one click

Expiry Monitor - See token status (valid/expired)

Time Machine - Saves last 15 API visits

Response Diff - Compare API versions side-by-side

Change Highlighting - Green (added), Red (removed), Yellow (modified)

What Json Formatter extensions you use?

deepesh verma — Fri, 27 Mar 2026 13:07:35 +0000

I am curious what type of extension you guys use as frontend developer