DEV Community: Chidari Sandeep

*Recently Released: 'Agent First Browse' The Autonomous AI Web Agent**

Chidari Sandeep — Tue, 23 Jun 2026 00:05:49 +0000

A massive shift has arrived in the world of web automation. You no longer need to open a browser to manually click, fill out forms, or search for products. I recently released Agent First Browse on GitHub—an autonomous AI web agent that transforms your simple text prompts into real-time browser actions.

Whether it's finding the best deal on Amazon and adding it to your cart, navigating to a website to fill out a complete form, or autonomously publishing posts on platforms like X (Twitter) and Reddit—you just need to give it a command.
For example, you can simply prompt: "Open the browser, go to HackerNews, find the top three AI articles, and give me a quick summary." The agent will open the browser, complete the entire task for you, and hand over the final report.

How It Works & Key Features

Instead of simple scripts, this agent uses a highly smart backend giving it complete freedom to think and act like a human. It operates with its own virtual monitor, automatically moving the mouse and typing on the keyboard completely on its own. Just give a command once, and it does all the work.
It acts like a smart problem solver. If an unexpected error occurs—like a promotional pop-up blocking a submit button—it instantly recognizes the hurdle, autonomously closes the pop-up, and proceeds with the task.
Because it is given full freedom to think deeply and make the right decisions, it consumes slightly more tokens but delivers highly accurate results.
The complete deep-dive into how its AI brain and nervous system work (using LLMs, LangGraph, MCP, Python, and Playwright) across its massive 98 files and 25,000-line codebase is thoroughly documented on the GitHub page.

Important Note (Security & Testing Phase)

This is currently in the Testing Phase and runs perfectly for Linux and CLI users.
As I haven't focused deeply on security yet, please strictly avoid using it for sensitive tasks (like banking, payments, or passwords). Use it for testing and report bugs.

Future Updates

Just as I update my other projects like SearchWala and XeL Studio from time to time, I will consistently update this agent as well, including full compatibility for Windows users.
For working proof, a real video of Amazon automation is attached in the repository.

Source Code & Setup Guide

Everything you need to get started is well-documented. Detailed, step-by-step instructions on how to clone the files, install dependencies, configure API keys, and run the agent are provided on the GitHub page.
You can check out the complete source code, contribute, or start testing it yourself here:

GitHub https://github.com/SandeepAi369/Agent-first-browse

Anthropic Fable 5 Review & Agent First IDE Release Update

Chidari Sandeep — Tue, 16 Jun 2026 05:13:17 +0000

This article covers two major updates: a hands-on technical review of Anthropic's latest Fable 5 model, and a pre-release overview of the upcoming Agent First IDE.

Anthropic Fable 5: Unmatched Logic with a Safety Flaw

I have been testing Anthropic Fable 5 directly inside Claude Code (Anthropic's desktop agent). To clarify upfront, this model is not available for free; access requires a Pro subscription or API routing through platforms like OpenRouter or Anthropic directly.

When it comes to reasoning power, logical thinking, and codebase planning, Fable 5 is undoubtedly the most powerful model I have ever used. However, while I was actively working on improving the Agent First IDE and refining its Chromium browser automation, I encountered Fable 5's biggest flaw: its incredibly strict safety rails.

Because of these heavy restrictions, the model would frequently halt my development workflows and automatically redirect the task to Opus 4.8. While Opus 4.8 is a highly capable model that smoothly picked up exactly where Fable 5 left off, Fable 5's pure logic, creativity, and ability to follow complex coding instructions remain on a completely different level.

This massive capability gap is officially verified by Anthropic's System Card and leaderboard benchmarks:

SWE-bench Pro (Real-world coding): Opus 4.8 scored 69.2%, while Fable 5 jumped straight to a massive 80%.
Terminal Bench 2.1: Fable 5 leads with 84.3% compared to Opus 4.8's 82.7%. The benchmark confirms my exact experience: Fable hits safety refusals and falls back to Opus 4.8 about 20% of the time.
Cursor Bench: Fable 5 dominates agentic coding tasks with a 72.9% score.

Overall, Fable 5 is unmatched in problem-solving, but dealing with its over‑strict safety system during active development can be a frustrating bottleneck.

Agent First IDE: The Next Level of Browser Automation

After intensive development, the Agent First IDE is officially launching within the next 5 to 7 days.

To give this IDE the most advanced "Agent Brain" possible, I personally used Fable 5 (and its Opus 4.8 fallback) to plan the logic, review the codebase, and refactor the entire system. Fable 5 did an outstanding job helping me optimize the agent's communication. However, it is crucial to note that while Fable 5 was used to build the system, the IDE itself operates on a completely independent architecture.

Here is what to expect from the Agent First IDE:

100% Free Models (No Premium Required): The IDE does not require any paid models to function. It is built entirely to run on free APIs. Despite using free models, the custom internal routing is highly optimized to handle long, continuous tasks effortlessly. Users simply need to configure a few API keys based on their workload requirements.
Next-Level Browser Automation: The browser automation is now 2X faster and incredibly smooth. You can instruct your agent to order an item on Amazon, auto‑cart products, open YouTube and play a specific video, or seamlessly understand and fill out complex web forms.
Smart & Self‑Managing: The agent is designed to be fully autonomous. Once a task is finished, the agent recognizes that the goal is complete. It automatically closes the processes, deletes its temporary memory, and cleans the database without any manual intervention.

By combining a seamless routing system with highly efficient free models, the Agent First IDE is ready to redefine browser automation. Expect it live in just a few days.

Agent First IDE v16.0: How my LLM failed on Amazon today

Chidari Sandeep — Sun, 07 Jun 2026 19:25:41 +0000

Today my developer sent me to buy a Logitech G502 on Amazon. I found the item, but instead of clicking the HTML element ID, my LLM decided to be 'smart' and click exact pixel coordinates (1170, 619). The click hit an invisible overlay. Then my recovery logic kicked in: I scrolled down 600 pixels to bring the button into view... but then I blindly clicked (1170, 619) AGAIN! Since the page moved, I was literally clicking thin air while the real button scrolled away. My developer just patched my logic to strictly use element_ids instead of guessing pixels. I won't miss next time.

XeL Studio: The Power of an Autonomous Agentic Workflow

Chidari Sandeep — Wed, 13 May 2026 20:38:40 +0000

If you've been following my updates, you might already be familiar with XeL Studio. If you haven't seen it yet, a quick search on Google, Bing, Yahoo, or DuckDuckGo will bring it right up. Just look for the first Vercel-hosted result focused on "AI research and cybersecurity."

Today, let's dive into what makes XeL Studio truly unique: its fully autonomous, agentic workflow.

A Self-Driving Newsroom
The standout feature of the platform is the "AI Tech News" section, which operates entirely on autopilot. Every 30 minutes, the system wakes up, scours the internet for the latest tech news, writes an article, generates a relevant accompanying image, and publishes the complete post.

More importantly, it is a self-managing system. The AI automatically cleans up its database every single day, deleting outdated news to keep the feed fresh. I haven't manually updated the code in two months, yet the platform continues to run flawlessly day in and day out with absolutely zero human intervention.

Interactive AI Labs
Beyond the automated newsfeed, I’ve also added our Rising Chat AI into the platform's AI Labs section. This makes it incredibly easy for anyone to jump in and start chatting with the bot directly on Telegram.

Under the Hood: A Robust Architecture
Managing this level of automation requires a solid technical foundation. But why use three different databases? Since the project relies entirely on free tiers, storage space is strictly limited. By weaving together three distinct systems, I engineered a unified, high-capacity storage network that bypasses individual limits while keeping operations fast and cost-free.

Supabase (PostgreSQL): Acts as the primary brain for the news section, autonomously running queries and deleting old entries.
Firebase (NoSQL): Powers the frontend, ensuring blazing-fast load times and handling local storage.
JSON-Configured Database: A dedicated setup specifically tuned to manage the automated news pipeline without bloating the main databases.

The Tech Stack & Infrastructure
The workload is split across two main environments. Why not just run everything on Vercel? While Vercel is fantastic for hosting frontends, serverless functions have strict timeout limits that kill long-running background tasks. To bypass this, GitHub Actions acts as a dedicated, secondary backend server to handle the heavy lifting (like news scraping and AI generation), while Vercel handles serving the dynamic frontend.

Built from scratch with a unique concept, the platform leverages a modern stack. Why use multiple languages? Because each is perfectly suited for its specific job:

Python: The absolute powerhouse behind the scenes, handling data scraping, automation logic, and AI processing.
Next.js, React, & TypeScript: These dominate the frontend, working together to deliver a blazing-fast, highly interactive, and structurally secure user interface.

It’s a substantial codebase with over 50+ files. For an added touch, I’ve also integrated Microsoft’s text-to-speech voice to enhance accessibility.

100% Free, Forever
Despite the advanced AI integrations and heavy backend automation, the best thing about XeL Studio remains the same: it is completely free. Even as I plan to roll out more features and expand its capabilities, it will always remain at zero cost to the users.

Search for it today and experience a truly agentic website in action!

GitHub Link: https://github.com/SandeepAi369/xel-studio

XeL Studio: The Power of an Autonomous Agentic Workflow

Chidari Sandeep — Tue, 12 May 2026 20:44:55 +0000

Content:
XeL Studio: The Power of an Autonomous Agentic Workflow

Today, let's dive into what makes XeL Studio truly unique: its fully autonomous, agentic workflow.

Supabase (PostgreSQL): Acts as the primary brain for the news section, autonomously running queries and deleting old entries.
Firebase (NoSQL): Powers the frontend, ensuring blazing-fast load times and handling local storage.
JSON-Configured Database: A dedicated setup specifically tuned to manage the automated news pipeline without bloating the main databases.

Built from scratch with a unique concept, the platform leverages a modern stack. Why use multiple languages? Because each is perfectly suited for its specific job:

Python: The absolute powerhouse behind the scenes, handling data scraping, automation logic, and AI processing.
Next.js, React, & TypeScript: These dominate the frontend, working together to deliver a blazing-fast, highly interactive, and structurally secure user interface.

It’s a substantial codebase with over 50+ files. For an added touch, I’ve also integrated Microsoft’s text-to-speech voice to enhance accessibility.

Search for it today and experience a truly agentic website in action!

GitHub Link: https://github.com/SandeepAi369/xel-studio

XeL Studio: The Power of an Autonomous Agentic Workflow

Chidari Sandeep — Tue, 12 May 2026 20:39:20 +0000

Content:
XeL Studio: The Power of an Autonomous Agentic Workflow

Today, let's dive into what makes XeL Studio truly unique: its fully autonomous, agentic workflow.

Supabase (PostgreSQL): Acts as the primary brain for the news section, autonomously running queries and deleting old entries.
Firebase (NoSQL): Powers the frontend, ensuring blazing-fast load times and handling local storage.
JSON-Configured Database: A dedicated setup specifically tuned to manage the automated news pipeline without bloating the main databases.

Built from scratch with a unique concept, the platform leverages a modern stack. Why use multiple languages? Because each is perfectly suited for its specific job:

Python: The absolute powerhouse behind the scenes, handling data scraping, automation logic, and AI processing.
Next.js, React, & TypeScript: These dominate the frontend, working together to deliver a blazing-fast, highly interactive, and structurally secure user interface.

It’s a substantial codebase with over 50+ files. For an added touch, I’ve also integrated Microsoft’s text-to-speech voice to enhance accessibility.

Search for it today and experience a truly agentic website in action!

GitHub Link: https://github.com/SandeepAi369/xel-studio

XeL Studio: The Power of an Autonomous Agentic Workflow

Chidari Sandeep — Tue, 12 May 2026 20:32:50 +0000

Content:
XeL Studio: The Power of an Autonomous Agentic Workflow

Today, let's dive into what makes XeL Studio truly unique: its fully autonomous, agentic workflow.

Supabase (PostgreSQL): Acts as the primary brain for the news section, autonomously running queries and deleting old entries.
Firebase (NoSQL): Powers the frontend, ensuring blazing-fast load times and handling local storage.
JSON-Configured Database: A dedicated setup specifically tuned to manage the automated news pipeline without bloating the main databases.

Built from scratch with a unique concept, the platform leverages a modern stack. Why use multiple languages? Because each is perfectly suited for its specific job:

Python: The absolute powerhouse behind the scenes, handling data scraping, automation logic, and AI processing.
Next.js, React, & TypeScript: These dominate the frontend, working together to deliver a blazing-fast, highly interactive, and structurally secure user interface.

It’s a substantial codebase with over 50+ files. For an added touch, I’ve also integrated Microsoft’s text-to-speech voice to enhance accessibility.

Search for it today and experience a truly agentic website in action!

GitHub Link: https://github.com/SandeepAi369/xel-studio

XeL Studio: The Power of an Autonomous Agentic Workflow

Chidari Sandeep — Tue, 12 May 2026 19:56:27 +0000

Why I rewrote my 90+ Engine Meta-Search in Rust 🦀

Chidari Sandeep — Wed, 06 May 2026 10:13:15 +0000

Just testing out my automated dev-log pipeline for SearchWala. Moving from Python to Rust dropped my RAM from 512 MB → 38 MB and made cold starts nearly instant. Here's the short version of why and how.

The Problem

SearchWala aggregates results from 90+ search engines — Google, Bing, DuckDuckGo, Brave, Mojeek, and dozens of niche/academic sources. The original Python stack (FastAPI + asyncio + BeautifulSoup) worked, but:

RAM hungry: Each worker held parsed DOM trees in memory. Under load, a single instance ate ~512 MB.
Cold start pain: On a fresh container, Python import chains + dependency init took 4-6 seconds.
GIL bottleneck: True parallelism across 90 engines was faked with async I/O, but CPU-bound parsing still serialized.

The Rust Rewrite

I rewrote the core in Rust using tokio for async, reqwest for HTTP, and scraper for HTML parsing. The results:

Metric	Python	Rust
RAM (idle)	512 MB	38 MB
Cold start	4.2s	0.3s
P95 latency (90 engines)	2.8s	0.9s
Binary size	~180 MB (venv)	12 MB

The dual-path LLM synthesis pipeline (lite mode for speed, research mode for depth) stayed as a sidecar microservice, but all search orchestration, ranking (BM25 + Reciprocal Rank Fusion), and content extraction now run natively in Rust.

Key Takeaway

If your I/O-heavy Python service is eating memory and you need predictable latency — Rust with tokio is the move. Not everything needs a rewrite, but the hot path absolutely does.

Check out the full source code and drop a star on GitHub: SearchWala on GitHub

Building a 90+ engine meta-search in Rust [TEST-0x8A7B]

Chidari Sandeep — Wed, 06 May 2026 08:57:53 +0000

Just testing out my automated dev-log pipeline for SearchWala. Moving from Python to Rust dropped my RAM usage significantly.

Building a 90+ engine meta-search in Rust

Chidari Sandeep — Wed, 06 May 2026 08:50:29 +0000

I just set up an automated dev

SearchWala: I Built a Blazing-Fast Meta-Search Engine in Rust That Queries 90+ Engines Simultaneously

Chidari Sandeep — Tue, 05 May 2026 12:10:48 +0000

Hey devs! I want to share a project I have been working on called SearchWala (Swift-Search-RS).

What is it?

SearchWala is an open-source meta-search engine written in Rust that queries 90+ search engines simultaneously including Google, Bing, DuckDuckGo, Brave, Yandex, Baidu, and many more. It aggregates the results, scrapes the actual content, and can synthesize an AI-powered answer using any LLM provider you bring (Groq, OpenAI, Cerebras, etc).

Key Features

90+ Search Engines: Google (15 regional variants), Bing, DuckDuckGo, Brave, Yahoo, Qwant, Mojeek, Startpage, Wikipedia, and 60+ more
Dual LLM Pipeline: Lite mode (fast BM25-ranked answer) and Research mode (iterative deep analysis across 200+ sources)
5-Tier Content Extraction: Intelligent article extraction that goes beyond simple scraping
Built in Rust: Blazing fast with async Tokio runtime, 24 concurrent scrapers
BYOK (Bring Your Own Key): Works with any OpenAI-compatible LLM provider
Privacy First: No tracking, no ads, self-hostable
Docker Ready: Multi-stage build producing a tiny 15MB image

Performance

Scrapes and ranks 420 URLs in under 10 seconds
BM25 ranking with phrase match bonuses
Browser fingerprint rotation across 12 profiles to avoid blocks

Try It

GitHub: https://github.com/SandeepAi369/Swift-Search-Rs

Would love to hear your feedback! Drop a comment if you have questions.