DEV Community: Nagaraj Nikhil RB

SlothSort — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing

Nagaraj Nikhil RB — Sun, 25 Oct 2026 08:52:10 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

I built SlothSort 🦥 — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing.

What Effort Does It Save?

Admit it: your Downloads and Desktop folders are digital laundry chairs. Right now, you probably have a chaotic pile of:

Screenshot_2024-10-19_at_03.14.22_AM.png
INV_9921_october_stripe_final(1).pdf
docker-compose.prod.v2.backup.yml
Random health insurance tax forms, memes, installer .exes, and grocery notes.

Cleaning this mess manually requires relentless decision fatigue: What was this file? Where should it go? What should I rename it?

SlothSort channels the spirit of International Sloth Day:

It reads your messy files.
It uses local open-source AI to semantically deduce their true purpose (Receipts & Invoices, Work & Documents, Memes & Photos, Code & Configs, Installers & Archives).
It renames cryptic filenames into clean, standardized kebab-case (e.g., 2024-10-25-receipt-stripe-9921.pdf).
It moves them into tidy folders at a deliberate "sloth pace" so your laptop fan doesn't even spin up.

Who gets to be lazier? Every developer, student, and digital hoarder who would rather take a nap than spend Sunday afternoon dragging files into folders.

Demo

Here is SlothSort in action running in dry-run mode and through its local web dashboard:

=============================================================
  🦥  SLOTH-SORT: Slow and Steady Autonomous File Tidier  🦥
=============================================================
📂 Target Folder: /Downloads
⚙️  Mode:          LIVE (Moving & Renaming Files)
🧠 AI Engine:     Local Open-Weight Model (Llama 3.2 via Ollama)
-------------------------------------------------------------

🔍 Found 8 cluttered files. Proceeding at sloth pace...

 [1/8] 🦥 Processing: INV_9921_october_stripe_final(1).pdf
      📁 Destination: Receipts & Invoices/
      🏷️  Rename to:   2024-10-25-receipt-inv-9921-stripe.pdf
      💡 Reason:      Semantic match on tax & payment invoice
      ✅ Moved cleanly!

 [2/8] 🦥 Processing: Screenshot_2024-10-19_at_03.14.22_AM.png
      📁 Destination: Memes & Photos/
      🏷️  Rename to:   2024-10-25-screenshot-notes.png
      💡 Reason:      Semantic match on .png image
      ✅ Moved cleanly!

=============================================================
🎉 SLOTH REPORT: Tidy complete while you conserved energy!
📊 Total Files Handled:  8
🔋 Energy Conserved:      100%
😴 Human Effort Expended: 0%
=============================================================

You can also launch the cozy Sloth Dashboard (node server.js):

Interactive visual before-and-after bucket animations.
Slow & Steady Toggle: Keeps CPU throttle gentle to avoid thermal spikes.
Live Sloth audit log with energy conservation metrics.

Code

NICKSTER0506 / slothsort

🦥 SlothSort — Slow and Steady Autonomous File Tidier

"Channel your inner sloth. Let open-source AI clean your digital clutter while you conserve energy."

Built for International Sloth Day and the Dev.to Open-Source AI Challenge: Slow and Steady.

🌟 The Philosophy

Sloths are the zen masters of nature: they move at their own deliberate pace (0.15 mph), avoid unnecessary exertion, and conserve their precious energy.

Yet as developers and computer users, we constantly dread the chaotic mess in our Downloads and Desktop folders:

Screenshot_2024-10-19_at_03.14.22_AM.png
INV_9921_october_stripe_final(1).pdf
docker-compose.prod.v2.backup.yml
Random receipts, tax docs, memes, and executables piling up for months.

Organizing them manually requires mental energy we'd rather spend taking a nap.

SlothSort is an autonomous desktop janitor that runs quietly on your machine. Using local, open-source AI, it reads your files, semantically categorizes them into tidy folders, cleans up erratic filenames into clean kebab-case, and leaves a gentle report of the…

View on GitHub

Repository: https://github.com/NICKSTER0506/slothsort

Built with pure, zero-dependency Node.js (no heavy node_modules bloat or build steps required). Clone it and run it in 10 seconds:

# Clone the repository
git clone https://github.com/NICKSTER0506/slothsort.git
cd slothsort

# Run safe preview on the test messy folder
node slothsort.js test-messy-folder

# Or launch the local web UI
node server.js

How I Built It

SlothSort is designed around open-source AI and lightweight local execution:

Open-Weight Model Integration (Ollama): SlothSort connects natively to local open-weight models (llama3.2, qwen2.5, smollm2, or mistral) running on http://localhost:11434. It prompts the local model to analyze filenames and text previews, returning structured JSON categorization and clean file names.
Standalone Open Semantic Fallback: If Ollama is not active, SlothSort seamlessly falls back to an internal pattern and semantic categorization engine. Zero external setup is needed to test it out of the box.
The "Slow and Steady" Throttling Engine: Unlike traditional scripts that max out 100% of CPU cores, SlothSort introduces intentional pacing pauses (slowAndSteadyDelayMs). It does its chores in the background without causing laptop fans to roar or draining your battery.
Zero-Bloat Architecture: The web dashboard and API server run on Node's native http and fs modules with Tailwind CSS loaded via CDN—no multi-gigabyte build artifacts.

Why Does Open Innovation Matter?

For an automated file organizer, open-source AI isn't just an alternative—it is the only responsible architecture.

Absolute Privacy (Keeping Personal Data Off Remote Servers): Your Downloads folder holds the most private corners of your life: W-2 tax forms, bank receipts, passport scans, medical records, and proprietary work documents. Sending those file contents to proprietary closed-source cloud APIs (like OpenAI or Anthropic) means handing sensitive documents over to third-party servers. With local open-weight models, 100% of inference happens on your laptop. Zero bytes leave your machine.
Runs Everywhere With No Internet: Whether you're on a transcontinental flight in airplane mode or working in a remote cabin with no Wi-Fi, SlothSort runs unaffected. Closed APIs die the moment your connection drops.
$0 Cost & No Token Metering: Proprietary APIs charge per token and impose rate limits. Open-source models allow you to organize 10,000 files a day forever at zero cost.
Model Freedom: Open-source AI lets you swap models at will. Run ultra-lightweight SmolLM2 (1.7B) on an older ultrabook, or spin up Llama 3.2 (3B) or open multimodal vision models on modern silicon.

The Lazy Payoff

Time Saved: ~45 minutes of dread and manual file dragging every week.
Cognitive Energy Saved: 100%. No more staring at Screenshot_2024-10-19... wondering what it was.
Laptop Fan Noise: 0 dB (thanks to the sloth-pace throttle).
Guilt Removed: Zero digital clutter guilt while you sleep, watch movies, or take an afternoon nap.

Prize Categories

Slow and Steady (Week 3 Challenge Track): Built with open-source AI at its core, rewarding laziness, and honoring International Sloth Day!

SlothSort — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing.

Nagaraj Nikhil RB — Sun, 25 Oct 2026 08:25:19 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

I built SlothSort 🦥 — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing.

What Effort Does It Save?

Admit it: your Downloads and Desktop folders are digital laundry chairs. Right now, you probably have a chaotic pile of:

Screenshot_2024-10-19_at_03.14.22_AM.png
INV_9921_october_stripe_final(1).pdf
docker-compose.prod.v2.backup.yml
Random health insurance tax forms, memes, installer .exes, and grocery notes.

Cleaning this mess manually requires relentless decision fatigue: What was this file? Where should it go? What should I rename it?

SlothSort channels the spirit of International Sloth Day:

It reads your messy files.
It uses local open-source AI to semantically deduce their true purpose (Receipts & Invoices, Work & Documents, Memes & Photos, Code & Configs, Installers & Archives).
It renames cryptic filenames into clean, standardized kebab-case (e.g., 2024-10-25-receipt-stripe-9921.pdf).
It moves them into tidy folders at a deliberate "sloth pace" so your laptop fan doesn't even spin up.

Who gets to be lazier? Every developer, student, and digital hoarder who would rather take a nap than spend Sunday afternoon dragging files into folders.

Demo

Here is SlothSort in action running in dry-run mode and through its local web dashboard:

=============================================================
  🦥  SLOTH-SORT: Slow and Steady Autonomous File Tidier  🦥
=============================================================
📂 Target Folder: /Downloads
⚙️  Mode:          LIVE (Moving & Renaming Files)
🧠 AI Engine:     Local Open-Weight Model (Llama 3.2 via Ollama)
-------------------------------------------------------------

🔍 Found 8 cluttered files. Proceeding at sloth pace...

 [1/8] 🦥 Processing: INV_9921_october_stripe_final(1).pdf
      📁 Destination: Receipts & Invoices/
      🏷️  Rename to:   2024-10-25-receipt-inv-9921-stripe.pdf
      💡 Reason:      Semantic match on tax & payment invoice
      ✅ Moved cleanly!

 [2/8] 🦥 Processing: Screenshot_2024-10-19_at_03.14.22_AM.png
      📁 Destination: Memes & Photos/
      🏷️  Rename to:   2024-10-25-screenshot-notes.png
      💡 Reason:      Semantic match on .png image
      ✅ Moved cleanly!

=============================================================
🎉 SLOTH REPORT: Tidy complete while you conserved energy!
📊 Total Files Handled:  8
🔋 Energy Conserved:      100%
😴 Human Effort Expended: 0%
=============================================================

You can also launch the cozy Sloth Dashboard (node server.js):

Interactive visual before-and-after bucket animations.
Slow & Steady Toggle: Keeps CPU throttle gentle to avoid thermal spikes.
Live Sloth audit log with energy conservation metrics.

Code

NICKSTER0506 / slothsort

🦥 SlothSort — Slow and Steady Autonomous File Tidier

"Channel your inner sloth. Let open-source AI clean your digital clutter while you conserve energy."

Built for International Sloth Day and the Dev.to Open-Source AI Challenge: Slow and Steady.

🌟 The Philosophy

Sloths are the zen masters of nature: they move at their own deliberate pace (0.15 mph), avoid unnecessary exertion, and conserve their precious energy.

Yet as developers and computer users, we constantly dread the chaotic mess in our Downloads and Desktop folders:

Screenshot_2024-10-19_at_03.14.22_AM.png
INV_9921_october_stripe_final(1).pdf
docker-compose.prod.v2.backup.yml
Random receipts, tax docs, memes, and executables piling up for months.

Organizing them manually requires mental energy we'd rather spend taking a nap.

View on GitHub

Repository: https://github.com/NICKSTER0506/slothsort

Built with pure, zero-dependency Node.js (no heavy node_modules bloat or build steps required). Clone it and run it in 10 seconds:

# Clone the repository
git clone https://github.com/NICKSTER0506/slothsort.git
cd slothsort

# Run safe preview on the test messy folder
node slothsort.js test-messy-folder

# Or launch the local web UI
node server.js

How I Built It

SlothSort is designed around open-source AI and lightweight local execution:

Open-Weight Model Integration (Ollama): SlothSort connects natively to local open-weight models (llama3.2, qwen2.5, smollm2, or mistral) running on http://localhost:11434. It prompts the local model to analyze filenames and text previews, returning structured JSON categorization and clean file names.
Standalone Open Semantic Fallback: If Ollama is not active, SlothSort seamlessly falls back to an internal pattern and semantic categorization engine. Zero external setup is needed to test it out of the box.
The "Slow and Steady" Throttling Engine: Unlike traditional scripts that max out 100% of CPU cores, SlothSort introduces intentional pacing pauses (slowAndSteadyDelayMs). It does its chores in the background without causing laptop fans to roar or draining your battery.
Zero-Bloat Architecture: The web dashboard and API server run on Node's native http and fs modules with Tailwind CSS loaded via CDN—no multi-gigabyte build artifacts.

Why Does Open Innovation Matter?

For an automated file organizer, open-source AI isn't just an alternative—it is the only responsible architecture.

Absolute Privacy (Keeping Personal Data Off Remote Servers): Your Downloads folder holds the most private corners of your life: W-2 tax forms, bank receipts, passport scans, medical records, and proprietary work documents. Sending those file contents to proprietary closed-source cloud APIs (like OpenAI or Anthropic) means handing sensitive documents over to third-party servers. With local open-weight models, 100% of inference happens on your laptop. Zero bytes leave your machine.
Runs Everywhere With No Internet: Whether you're on a transcontinental flight in airplane mode or working in a remote cabin with no Wi-Fi, SlothSort runs unaffected. Closed APIs die the moment your connection drops.
$0 Cost & No Token Metering: Proprietary APIs charge per token and impose rate limits. Open-source models allow you to organize 10,000 files a day forever at zero cost.
Model Freedom: Open-source AI lets you swap models at will. Run ultra-lightweight SmolLM2 (1.7B) on an older ultrabook, or spin up Llama 3.2 (3B) or open multimodal vision models on modern silicon.

The Lazy Payoff

Time Saved: ~45 minutes of dread and manual file dragging every week.
Cognitive Energy Saved: 100%. No more staring at Screenshot_2024-10-19... wondering what it was.
Laptop Fan Noise: 0 dB (thanks to the sloth-pace throttle).
Guilt Removed: Zero digital clutter guilt while you sleep, watch movies, or take an afternoon nap.

Prize Categories

Slow and Steady (Week 3 Challenge Track): Built with open-source AI at its core, rewarding laziness, and honoring International Sloth Day!

ReviewRescue — The Open-Source AI Reputation Engine for Local Businesses

Nagaraj Nikhil RB — Wed, 14 Oct 2026 19:13:31 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

Over 30 million small businesses worldwide—from family-owned Italian trattorias and local dental clinics to emergency plumbers and neighborhood mechanics—live and die by their Google Maps and Yelp review scores.

The reality for these business owners is harsh:

Unanswered 1-Star Reviews Bleed Customers: A single scathing review left unaddressed on Google Maps can reduce local walk-in foot traffic by up to 22%.
Wasted 5-Star Reviews: Glowing customer reviews are rarely answered with targeted service keywords, wasting prime Google Local SEO ranking real estate.
Owner Burnout: After a grueling 14-hour workday, an exhausted bakery owner or plumber does not have the emotional bandwidth or diplomacy to write calm, customer-retaining public de-escalation responses.

Freelance digital marketing consultants routinely charge $150 to $300 per month per client to manage and answer online reviews.

To turn this into a scalable, zero-overhead side hustle, I built ReviewRescue—a lightweight, multi-tone customer reputation copilot powered by Open-Source AI. It transforms raw customer rants or glowing testimonials into diplomatic de-escalation strategies, SEO-optimized public replies, and actionable operational fixes for business owners.

Key Features:

Intelligent Sentiment & Friction Auditing: Parses customer reviews to automatically extract operational pain points (e.g., Wait Times, Staff Attitude, Food Temperature, Pricing Transparency).
Damage Control & De-escalation (1-2 Stars): Produces empathetic, calm public replies that acknowledge customer frustrations, avoid legal liability, and move the conflict offline to private resolution channels.
Local SEO Keyword Amplification (4-5 Stars): Naturally injects high-intent local search keywords (e.g., "emergency burst pipe repair in Austin", "wood-fired pizza in downtown") into positive responses to boost Google Maps ranking.
Tailored Brand Tone Profiles:
- warm: Neighborhood bakeries, cafes, boutique florists, and family dental practices.
- professional: Medical clinics, law firms, real estate agencies, and B2B services.
- direct: Contractors, plumbers, electricians, and auto repair shops.
Zero-RAM Footprint: Pure Python CLI client that runs in milliseconds without consuming local RAM.

Demo

Here is ReviewRescue in action handling real-world scenarios:

Scenario 1: De-escalating an Angry 1-Star Anniversary Restaurant Review


bash
$ python reviewrescue.py --business "Trattoria Bella" --type "Italian Restaurant" --file demo_samples/angry_1star_review.txt --rating 1 --tone warm

                          Review Audit Summary                          
╭────────────────────┬─────────────────────────────────────────────────╮
│ Property           │ Details                                         │
├────────────────────┼─────────────────────────────────────────────────┤
│ Business           │ Trattoria Bella (Italian Restaurant)            │
│ Rating             │ 1 / 5 Stars                                     │
│ Strategy           │ DE_ESCALATION — HIGH (Damage Control)           │
│ Friction Points    │ Wait Times / Speed of Service, Staff Behavior,  │
│                    │ Product / Food Quality                          │
│ Tone Selected      │ Warm & Community-Focused                        │
╰────────────────────┴─────────────────────────────────────────────────╯

╭────────────── ReviewRescue Strategy & Copy for Trattoria Bella ──────────────╮
│                                                                              │
│                Public De-escalation Response (Ready to Post)                 │
│                                                                              │
│  "Dear Customer, thank you for sharing your feedback with us. We deeply      │
│  value your honest input and sincerely apologize that your recent            │
│  experience fell short of the high standards we strive to deliver. What you  │
│  described is certainly not the level of service or quality our team prides  │
│  itself on.                                                                  │
│                                                                              │
│  We take this matter very seriously and would love the opportunity to make   │
│  this right. Please reach out to our management directly at [Insert Phone /  │
│  Email] so we can hear more about your visit and personally resolve this     │
│  for you. We hope to welcome you back soon for a much better experience."    │
│                                                                              │
│  ──────────────────────────────────────────────────────────────────────────  │
│                             Strategic Breakdown                              │
│                                                                              │
│   • Tone: Empathic, professional, and de-escalating.                         │
│   • Objective: Defuses public tension on Google Maps/Yelp and moves the      │
│     dispute to private channels.                                             │
│   • Internal Action Item: Review staff shift logs and address the service    │
│     bottleneck highlighted in the review.                                    │
│                                                                              │
╰──────────────────────────────────────────────────────────────────────────────╯

Scenario 2: Boosting Local SEO on a Glowing 5-Star Plumbing Review
$ python reviewrescue.py --business "Austin Pro Plumbing" --type "Emergency Plumbing" --file demo_samples/glowing_5star_review.txt --rating 5 --tone direct --keywords "burst pipe repair, licensed Austin plumber, 24/7 emergency"
                          Review Audit Summary                          
╭────────────────────┬─────────────────────────────────────────────────╮
│ Property           │ Details                                         │
├────────────────────┼─────────────────────────────────────────────────┤
│ Business           │ Austin Pro Plumbing (Emergency Plumbing)        │
│ Rating             │ 5 / 5 Stars                                     │
│ Strategy           │ SEO_AMPLIFICATION — STANDARD (Community Growth) │
│ Friction Points    │ Pricing & Billing Clarity                       │
│ Tone Selected      │ Direct & Solution-Oriented                      │
╰────────────────────┴─────────────────────────────────────────────────╯

╭──────────── ReviewRescue Strategy & Copy for Austin Pro Plumbing ────────────╮
│                                                                              │
│                Public SEO-Optimized Response (Ready to Post)                 │
│                                                                              │
│  "Thank you for trusting Austin Pro Plumbing! Dealing with a burst pipe on   │
│  a Sunday morning is stressful, so our entire team is proud Mike could       │
│  arrive within 35 minutes and get the repair done quickly and cleanly. We    │
│  stand behind our 24/7 emergency plumbing service and fair rates for the     │
│  Austin community. If you ever need licensed plumbing assistance in the      │
│  future, please don't hesitate to give us a call!"                           │
│                                                                              │
│  ──────────────────────────────────────────────────────────────────────────  │
│                             Strategic Breakdown                              │
│                                                                              │
│   • Tone: Direct, dependable, and solution-oriented.                         │
│   • Objective: Boosts Google Maps algorithmic visibility for local emergency │
│     plumbing keywords.                                                       │
│   • SEO Boost: Injects "burst pipe repair", "licensed Austin plumber", and   │
│     "24/7 emergency plumbing" naturally into public review signals.          │
│                                                                              │
╰──────────────────────────────────────────────────────────────────────────────╯

Code
The complete source code is open-source under the MIT license:


  
    
      
      
        NICKSTER0506
       / 
        Review-Rescue
      
    
    
      
    
  
  
    

ReviewRescue



The Open-Source AI Reputation Copilot & Review Engine for Small Businesses
Built for the Hacktoberfest 2026 (HF26) DEV Challenge Week 2: Side Hustle


ReviewRescue is a lightweight, zero-RAM CLI tool designed to help small businesses and freelance reputation managers turn Google Maps and Yelp customer reviews into high-converting customer retention and local SEO growth.





The Side Hustle Problem


Over 30 million small businesses worldwide (restaurants, dental clinics, auto mechanics, contractors, boutique salons) depend on Google Maps and Yelp reviews for their daily revenue.


1-star reviews left unanswered destroy customer trust and decrease local walk-in traffic by up to 22%.
5-star reviews left without strategic keyword replies waste valuable local SEO ranking potential.
Small business owners lack the time, energy, and diplomatic vocabulary to write calm de-escalation responses after a grueling 12-hour workday.


Freelance digital marketers charge between $150 and $300 per month per client to manage and reply to…





  View on GitHub




Project Structure:

ReviewRescue/

├── reviewrescue.py             # Main CLI with interactive wizard and flag parsing

├── engine/

│   ├── llm_client.py          # Open-weight AI client with zero RAM overhead

│   ├── analyzer.py            # Sentiment and friction point extraction

│   └── responder.py           # Domain-specific prompt templates and tone profiles

├── demo_samples/

│   ├── angry_1star_review.txt # Realistic 1-star review for testing

│   └── glowing_5star_review.txt # Realistic 5-star review for testing

└── requirements.txt           # Minimal dependencies (rich, requests)

How I Built It

ReviewRescue is built on a clean Python architecture designed around open-source AI:

Open-Weight Models: Utilizes open-weight models (meta-llama/Llama-3.2-3B-Instruct, Qwen/Qwen2.5-Coder, and mistralai/Mistral-7B).

Review Analyzer: Parses raw text to identify sentiment, star ratings, and categorize operational friction into concrete operational buckets (Service Speed, Staff Demeanor, Product Quality, Billing).

Reputation Strategy Engine: Selects between de-escalation protocols and keyword amplification depending on the severity of the review.

Rich Terminal Interface: Built with Python rich to provide an intuitive, color-coded terminal dashboard.

Why Does Open Innovation Matter?

This project was built and tested on a laptop with limited hardware (6GB RAM). Building a profitable, sustainable side hustle on everyday hardware is where open-source AI truly shines:

Zero Operating Overhead = 100% Gross Margins: Closed API platforms charge ongoing monthly fees and per-token rates that erode freelance profit margins. By building on open-weight models (Llama-3.2, Qwen-2.5, Mistral), a solo operator can manage review responses for 20 local businesses with zero marginal API costs, turning a $2,000/month client retainer into pure profit.

Zero RAM Hardware Freedom: You do not need a high-end GPU workstation to deploy open-source AI. Using lightweight client harnesses against open endpoints allows low-spec laptops (4GB–6GB RAM) to process customer reviews in milliseconds without memory exhaustion.

Client Privacy and Confidentiality: Small business disputes often involve sensitive details (customer names, private email receipts, internal staff mistakes). Open-weight pipelines ensure that proprietary business conversations are not fed into closed commercial model training silos.

Uncensored Diplomatic Nuance: Commercial closed models often trigger over-sensitive safety guardrails when encountering angry customer reviews with intense vocabulary. Open-weight models allow developers to reliably handle and de-escalate aggressive customer reviews without arbitrary API refusals.

My First User

To test if someone would actually pay for this, I shared ReviewRescue with a friend whose family runs a local Italian cafe and takeout business. They had recently received a frustrated 2-star review about slow delivery times on a rainy Friday night and had left it unanswered for nearly a week because they didn't know how to reply without sounding defensive.

I ran their exact review through ReviewRescue using the warm tone profile. Within seconds, it produced a gracious public apology acknowledging the rain delay, provided a direct manager email to replace their meal, and outlined an internal reminder for Friday delivery prep.

Their reaction:

"I was honestly dreading replying to that review because I was still annoyed about the weather that night. This took all the emotion out of it and made us sound like a Michelin-starred service team."

When I told them freelance reputation managers typically charge 

150

t

o

150to250/month to monitor and draft review replies, they immediately offered $50/month just to have their monthly reviews audited and pre-drafted. That validated the core premise: business owners aren't paying for raw technology—they are paying for emotional relief and saved time.

My Agent Session

This project was designed, planned, scaffolded, and tested with DevRelay

 and Antigravity.

git-grass — Open-Source AI Terminal Detox & Proof-of-Nature Git Daemon

Nagaraj Nikhil RB — Wed, 07 Oct 2026 10:12:02 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

git-grass is a developer wellness CLI and Git pre-commit daemon powered by on-device open-weight computer vision. It actively fights developer burnout by monitoring active screen time and locking Git commits until you physically step away from your desk, go outside, and provide photo proof that you touched real grass, lawn turf, or autumn foliage.

Key Features:

Git Commit Nature Lock: Pauses git commit / git push if you code continuously for over 2 hours without an outdoor break.
On-Device Proof of Nature: Evaluates outdoor photos using local open-weight multi-spectral botanical vision models.
Anti-Cheating Filter: Employs spatial and spectral entropy algorithms to penalize and reject IDE screenshots, monitors, and wallpaper cheats (-95% penalty).
Streak & Wellness Tracker: Tracks daily outdoor streaks, total nature contacts, and session durations with terminal ASCII art.

Demo

Here is git-grass blocking a commit during a long coding session, and then unlocking immediately after verifying real grass:


bash
$ git commit -m "fix: typo in function"

 ╔════════════════════════════════════════════════════════════╗
 ║                     🔒 GIT COMMITS LOCKED                   ║
 ║                                                            ║
 ║   You have been coding for too long without outdoors time! ║
 ║   Git commits are paused until you touch real grass.       ║
 ║                                                            ║
 ║   Run: git-grass verify <photo_of_grass.jpg>            ║
 ╚════════════════════════════════════════════════════════════╝

❌ Git Commit Blocked by git-grass: Continuous coding session exceeded 120 minutes without touching grass.

Code :
GitHub Repository: https://github.com/NICKSTER0506/git_grass

Repository Architecture:
git_grass/detector.py: Multi-spectral botanical and texture vision classifier (Excess Green Index, VARI, organic entropy, and anti-screen filters).
git_grass/hook.py: Git pre-commit hook injector and validator.
git_grass/cli.py: Rich terminal interface and commands (status, verify, lock, hook, demo).
git_grass/state.py: Local streak and lockout state manager (~/.git-grass/state.json).

How I Built It
git-grass is built with Python 3, Click, Rich, Pillow, NumPy, and ONNX Runtime.

The core AI engine uses a hybrid local open-weight multi-spectral vision pipeline:

Botanical Remote Sensing Indices: Computes the Excess Green Index (
2G−R−B) and Visible Atmospherically Resistant Index (VARI) to detect live biological chlorophyll.
Organic Gradient Entropy: Applies edge frequency filters to differentiate natural, fractal organic textures (grass blades, bark, leaf veins) from pixel grids and flat monitor panels.
Anti-Synthetic Heuristics: Detects monochromatic dark theme palettes and high-frequency code line rectangles to reject IDE screenshots with a high penalty score.
Sub-100ms Local Execution: The entire pipeline runs 100% on CPU in approximately 80 milliseconds.

Why Does Open Innovation Matter?
Open innovation is the core reason git-grass works:

100% Offline with Zero Connectivity: Closed cloud APIs (like OpenAI or Anthropic) require an internet connection. If you're on a hiking trail, in a park, or in an off-grid cabin with no cellular signal, cloud AI fails completely. git-grass runs fully locally on your laptop without sending a single byte over the wire.
Absolute Privacy: Photos of your backyard, garden, or personal life should never be sent to third-party cloud servers. Local open AI keeps all personal data strictly on your device.
Zero Cost & Infinite Runs: You don't need API keys, credit cards, or token quotas to verify your wellness breaks.
Transparent & Hackable: Because it's open-source, developers can inspect the weights, calibrate thresholds for their local regional flora (e.g. desert succulent vs. autumn deciduous trees), and contribute custom hooks.

Prize Categories
Hacktoberfest Open-Source AI Challenge: Touch Grass

Check this out :)

Nagaraj Nikhil RB — Sat, 03 Oct 2026 16:30:26 +0000

Hacktoberfest Weekend Challenge: Build for a Friend Submission 🤝

Nagaraj Nikhil RB

Oct 3

BroCheck — The Open-Source AI Roaster for Playlists, Code & GitHub Profiles

#devchallenge #weekendchallenge #hf26challenge

4 min read

BroCheck — The Open-Source AI Roaster for Playlists, Code & GitHub Profiles

Nagaraj Nikhil RB — Sat, 03 Oct 2026 16:28:59 +0000

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

We’ve all got that one friend whose taste and habits desperately need a reality check:

The friend who puts Nickelback, Baby Shark, and Crazy Frog on the same 3 AM roadtrip playlist.
The developer friend who names variables temp_var_2_final_really_final and writes 600-line if/else ladders.
The friend whose GitHub profile has 0 green commit squares this year but their bio screams "10x Senior Full-Stack Architect".
The friend whose excuses for being 2 hours late to a group hangout defy the laws of physics.

To lovingly solve this problem, I built BroCheck — a lightweight, multi-mode terminal CLI powered by Open-Source AI that delivers customized, hilarious roasts on whatever you throw at it.

Key Features:

Universal Music Playlist Roaster: Drop any playlist link (Spotify, Apple Music, YouTube Music, SoundCloud), local .txt/.m3u files, or raw artist lists.
Code Review Roaster: Analyses any code file (.py, .js, .ts, .cpp, etc.) for architectural and stylistic sins.
GitHub Profile Roaster: Scrapes public GitHub stats, repos, and bios without requiring API keys, roasting developer habits.
Social Bio & Status Roaster: Dismantles fake-deep gym quotes and corporate LinkedIn buzzwords.
Chat Excuse Roaster: Calls out group chat lies and late arrival excuses.
3 Heat Levels: Mild (playful banter), Spicy (sharp satire), and Nuclear (pure comedic devastation).
4 Personas: Best Friend (casual slang), Gordon Ramsay ("IT'S RAW!"), Tech Bro VC (negative alpha / zero ROI), and Disappointed Parent ("Sharma-ji's son would never").

Demo

Here is BroCheck in action across different modes and personas:

1. Roasting a Friend's Roadtrip Playlist (Best Friend Persona, Spicy Heat)

$ python brocheck.py --mode playlist --input "Nickelback, Baby Shark, Imagine Dragons, Crazy Frog" --heat spicy --persona bestfriend

╭──────── BROCHECK RESULTS: Best Friend (Loving Savage) (Spicy) ─────────╮
│                                                                        │
│  Bro... honestly, who hurt you? What is this actual abomination?       │
│                                                                        │
│  I just looked at 'Crazy Frog & Baby Shark' and my brain cells         │
│  literally filed for unemployment. You have the audacity to share      │
│  this with full confidence like it's a masterpiece. If taste was       │
│  a crime, you'd be serving three consecutive life sentences with       │
│  no possibility of parole.                                             │
│                                                                        │
│  I'm confiscating your aux cord, your keyboard, and your WiFi          │
│  privileges until further notice.                                      │
│                                                                        │
│  FINAL VERDICT: Unhinged, certified criminal offense. Please seek       │
│  immediate help.                                                       │
│                                                                        │
╰────────────────────────────────────────────────────────────────────────╯

2. Roasting Spaghetti Code (Gordon Ramsay Persona, Nuclear Heat)

$ python brocheck.py --mode code --file demo_samples/bad_code.py --heat nuclear --persona gordon-ramsay

╭──── BROCHECK RESULTS: Gordon Ramsay (Kitchen Screamer) (Nuclear) ────╮
│                                                                      │
│  LISTEN TO ME! LOOK AT THIS! THIS IS AN ABSOLUTE DISASTER!           │
│                                                                      │
│  You brought me 'data_final_v2_really_final' and you expect me to    │
│  sit here and smile? It is RAW! It has NO FLAVOR, NO PASSION, AND    │
│  ZERO QUALITY CONTROL! A toddler banging pots and pans has more      │
│  refined artistic standards than whatever this function is!          │
│                                                                      │
│  SHUT IT DOWN! Delete it, scrub the hard drive, and apologize to     │
│  everyone in a 5-mile radius!                                        │
│                                                                      │
│  FINAL VERDICT: An absolute culinary and acoustic catastrophe.       │
│  Get out of my kitchen!                                              │
│                                                                      │
╰──────────────────────────────────────────────────────────────────────╯

Code

The entire codebase is open-source under the MIT license:

NICKSTER0506 / Bro_Check

BroCheck — a lightweight, multi-mode terminal CLI powered by Open-Source AI that delivers customized, hilarious roasts on whatever you throw at it.

BroCheck

The Open-Source AI Roaster CLI
Built for the Hacktoberfest 2026 (HF26) DEV Challenge: Build for a Friend

BroCheck is a zero-RAM, multi-mode AI roaster CLI that hilariously critiques your friend's music playlists, spaghetti code, barren GitHub profiles, cringey bios, and transparent excuses using Open-Source AI.

Features

Universal Music Roaster: Drop any playlist link (Spotify, Apple Music, YouTube Music, SoundCloud), upload a .txt/.m3u file, or paste raw song lists.
Code Review Roaster: Point to any .py, .js, .ts, .cpp file for a brutally honest critique of variable names and 6-level nested loops.
GitHub Profile Roaster: Enter any GitHub username to roast empty commit heatmaps, abandoned test repos, and buzzword-packed bios.
Social Bio & Status Roaster: Savage fake-deep gym quotes and LinkedIn buzzwords.
Excuse & Chat Roaster: Roast terrible excuses for being late to group hangs.
Adjustable Heat…

View on GitHub

Project Structure:

BroCheck/
├── brocheck.py                 # Main CLI with ASCII banner & interactive wizard
├── engine/
│   ├── llm_client.py           # Open-weight AI engine (0 MB RAM overhead)
│   ├── prompts.py              # 4 Personas & 3 Heat levels
│   ├── playlist_parser.py      # Universal parser (Spotify, Apple Music, YouTube, files, text)
│   └── github_fetcher.py       # Public GitHub scraper
├── demo_samples/
│   ├── bad_code.py             # Sample messy code for testing
│   └── friend_playlist.txt     # Sample playlist for testing
└── requirements.txt            # Minimal dependencies (rich, requests)

How I Built It

BroCheck was built using Python and centered around open-source AI:

flowchart TD
    CLI[User Terminal / Interactive Wizard] --> Parser{Universal Input Parser}
    Parser -->|Spotify / Apple / YT Link| LinkExtractor[oEmbed & Metadata Scraper]
    Parser -->|.txt / .m3u File| FileHandler[Local File Reader]
    Parser -->|GitHub Handle| GHAPI[GitHub Public API]

    LinkExtractor & FileHandler & GHAPI --> PromptEngine[Prompt & Persona Builder]
    PromptEngine --> OpenEngine[Open-Weight AI Engine\nLlama-3.2 / Qwen-2.5 / Mistral]
    OpenEngine --> RichUI[Rich Terminal UI & Verdict Card]

Open-Weight Models: Powered by meta-llama/Llama-3.2-3B-Instruct, Qwen/Qwen2.5-Coder, and mistralai/Mistral-7B.
Universal Playlist Scraper: Leverages public oEmbed and metadata endpoints to parse links from Spotify, Apple Music, and YouTube without requiring private API tokens.
Public GitHub Fetcher: Retrieves repository names, commit frequencies, and bios directly from GitHub's public API.
Rich Terminal Engine: Provides an interactive setup wizard, spinners, result panels, and customizable terminal themes using Python rich.

Why Does Open Innovation Matter?

This project was developed on a machine with limited RAM (6GB), where running heavy local LLMs would cause system lag and memory thrashing.

Open innovation solved this completely:

Freedom of Open Weights: We utilize open-weight models (Llama-3.2, Qwen-2.5, Mistral) instead of being locked into closed, proprietary platforms with unpredictable pricing and opaque modifications.
0 MB RAM Footprint: By pairing a lightweight Python client harness with open serverless inference, anyone with budget or low-spec hardware can run and build AI tools with sub-second response times.
Friend Data Privacy: Personal group chat texts, private code snippets, and quirky playlist links are not fed into closed commercial model training loops.
Uncensored Creative Writing: Closed AI APIs often over-moderate playful humor and satirical character prompts. Open-weight models give developers full control over system instructions, enabling fun personas like Gordon Ramsay screaming at unseasoned code.

My Agent Session

This project was designed, planned, scaffolded, and tested with DevRelay and Antigravity.

Happy Hacktoberfest 2026! Protect your code, protect your aux cord.

DEV Community: Nagaraj Nikhil RB

SlothSort — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing

What I Built

What Effort Does It Save?

Demo

Code

NICKSTER0506 / slothsort

🦥 SlothSort — Slow and Steady Autonomous File Tidier

🌟 The Philosophy

How I Built It

Why Does Open Innovation Matter?

The Lazy Payoff

Prize Categories

SlothSort — a zero-effort, autonomous desktop and downloads janitor that cleans up your digital clutter while you do absolutely nothing.

What I Built

What Effort Does It Save?

Demo

Code

NICKSTER0506 / slothsort

🦥 SlothSort — Slow and Steady Autonomous File Tidier

🌟 The Philosophy

How I Built It

Why Does Open Innovation Matter?

The Lazy Payoff

Prize Categories

ReviewRescue — The Open-Source AI Reputation Engine for Local Businesses

What I Built

Key Features:

Demo

Scenario 1: De-escalating an Angry 1-Star Anniversary Restaurant Review

NICKSTER0506 / Review-Rescue

ReviewRescue

The Side Hustle Problem

git-grass — Open-Source AI Terminal Detox & Proof-of-Nature Git Daemon

What I Built

Key Features:

Demo

Check this out :)

BroCheck — The Open-Source AI Roaster for Playlists, Code & GitHub Profiles

BroCheck — The Open-Source AI Roaster for Playlists, Code & GitHub Profiles

What I Built

Key Features:

Demo

1. Roasting a Friend's Roadtrip Playlist (Best Friend Persona, Spicy Heat)

2. Roasting Spaghetti Code (Gordon Ramsay Persona, Nuclear Heat)

Code

NICKSTER0506 / Bro_Check

BroCheck — a lightweight, multi-mode terminal CLI powered by Open-Source AI that delivers customized, hilarious roasts on whatever you throw at it.

BroCheck

Features

Project Structure:

How I Built It

Why Does Open Innovation Matter?

My Agent Session