LibTrack AI

AI-powered dependency version tracker for software projects.
Register your projects, define your tech stack, and get automatically notified whenever a new version of a library, language, or tool you depend on is released — with AI-predicted future releases included.

---

What it does

Most teams find out about dependency updates too late — during a security audit, a failing build, or a colleague's offhand comment. LibTrack AI fixes that by:

• Continuously checking official registries (PyPI, npm, RubyGems, Cargo, GitHub) for the latest stable versions of every dependency in your stack
• Sending email notifications to your team when updates are available, filtered by type (major / minor / future)
• Using an LLM (Groq) backed by web search (Serper) to predict upcoming releases before they are officially published, with a confidence score per prediction
• Giving you a live dashboard with a health score, update breakdown charts, and a future roadmap view

---

Features

Feature	Description
Project Registry	Register any project with team names, emails, and a full technology stack (languages, libraries, tools, modules)
Multi-Registry Detection	Fetches latest stable versions from PyPI, npm, RubyGems, Cargo, and GitHub Releases
AI Fallback	If an official API fails, Serper (Google search) + Groq LLM extracts the correct version from the web
Future Version Prediction	Groq analyses release cadence and public signals to predict upcoming releases with confidence scores
Smart Notifications	Email alerts via Mailtrap — configurable per project (major only, minor, future, or all)
Notification Pause	Temporarily silence notifications for a project without removing it
Confidence Threshold	Only receive future-update notifications above a user-defined confidence level (0–100%)
Health Score	Dashboard metric showing what percentage of your tracked libraries are up-to-date
Dark / Light Theme	GitHub-dark design system with a persistent theme toggle

---

Tech Stack

Backend

• Django 5.2 — web framework
• Python 3.14
• SQLite — default database (drop-in replaceable with PostgreSQL)

AI / Data

• Groq API — LLM inference for version analysis and future release prediction
• Serper API — Google Search API used as a version-detection fallback and for future release signals

Notifications

• Mailtrap — transactional email (bulk API)

Frontend

• Bootstrap 5.3 — grid and utility classes
• Custom GitHub-dark design system (style.css) — full dark/light token palette, overrides all Bootstrap visuals
• Chart.js — dashboard bar and doughnut charts
• Bootstrap Icons

Scheduler

• schedule library — powers the daily check loop inside the run_daily_check management command

---

How the Daily Check Works

python manage.py run_daily_check
        │
        ▼
1. LibrarySyncService     — link all StackComponents to central Library records
        │
        ▼
2. VersionFetchService    — query official registries for latest stable versions
        │                   fallback: Serper search + Groq LLM extraction
        ▼
3. FutureUpdateService    — predict upcoming releases via Groq analysis
        │                   stores confidence score per prediction
        ▼
4. NotificationService    — send emails to project teams via Mailtrap
                            respects per-project type filter + confidence threshold

---

Getting Started

Prerequisites

• Python 3.11+
• API keys for Groq, Serper, and Mailtrap (all have free tiers)

1. Clone and install

git clone https://github.com/your-username/LibTrack-AI.git
cd LibTrack-AI
python -m venv venv
source venv/bin/activate        # Windows: venv\Scripts\activate
pip install -r requirements.txt

2. Configure environment

Create a .env file in the project root:

# Django
SECRET_KEY=your-secret-key-here
DEBUG=True
ALLOWED_HOSTS=127.0.0.1,localhost
TIME_ZONE=Asia/Kolkata

# API Keys
GROQ_API_KEY=your-groq-api-key
SERPER_API_KEY=your-serper-api-key

# Email (Mailtrap)
MAILTRAP_API_KEY=your-mailtrap-api-key
MAILTRAP_FROM_EMAIL=noreply@yourdomain.com

# Feature flags (optional — defaults shown)
LIBTRACK_ENABLE_EMAIL_NOTIFICATIONS=True
LIBTRACK_USE_OFFICIAL_APIS=True
LIBTRACK_FETCH_DEBUG_MODE=False
LIBTRACK_MIN_FUTURE_CONFIDENCE=40
LIBTRACK_MAX_FUTURE_CONFIDENCE=95
LIBTRACK_API_RATE_LIMIT_SECONDS=1.5
LIBTRACK_DEDUP_WINDOW_HOURS=24
LIBTRACK_EMAIL_BATCH_SIZE=10
LIBTRACK_MAX_NOTIFICATION_RETRIES=3
LIBTRACK_LOG_LEVEL=INFO

3. Run migrations and create a user

python manage.py migrate
python manage.py createsuperuser

4. (Optional) Seed demo data

python manage.py seed_test_data

5. Start the development server

python manage.py runserver

Open http://127.0.0.1:8000/libtracker/login/

---

Running the Daily Check

Run project:

python manage.py runserver

Clear cache:

python manage.py clear_cache

One-off run (useful for testing):

python manage.py run_daily_check --auto --time 22:22

Continuous scheduler (runs every 24 hours):

python manage.py run_daily_check

In production, run this as a background service via systemd, supervisor, or a cron job.

---

Useful Management Commands

Command	Purpose
run_daily_check	Full sync → fetch → predict → notify pipeline
clear_cache	Wipe UpdateCache and FutureUpdateCache tables
seed_test_data	Populate database with sample projects and stack data
test_email	Send a test email to verify Mailtrap configuration
verify_adapters	Check that all registry adapters are reachable

---

Supported Registries

Registry	Ecosystem
PyPI	Python packages
npm	JavaScript / Node.js
RubyGems	Ruby gems
Cargo (crates.io)	Rust crates
GitHub Releases	Any library with a public GitHub repo

---

Environment Variables Reference

Variable	Required	Default	Description
SECRET_KEY	Yes	—	Django secret key
DEBUG	No	False	Enable Django debug mode
ALLOWED_HOSTS	No	127.0.0.1	Comma-separated allowed hosts
TIME_ZONE	No	UTC	Django timezone
GROQ_API_KEY	Yes	—	Groq LLM API key
SERPER_API_KEY	Yes	—	Serper Google Search API key
MAILTRAP_API_KEY	Yes*	—	Mailtrap bulk API key
MAILTRAP_FROM_EMAIL	Yes*	—	Sender email address
LIBTRACK_ENABLE_EMAIL_NOTIFICATIONS	No	True	Toggle email dispatch globally
LIBTRACK_USE_OFFICIAL_APIS	No	True	Use official registry APIs first
LIBTRACK_MIN_FUTURE_CONFIDENCE	No	40	Minimum confidence % to store a future prediction
LIBTRACK_MAX_FUTURE_CONFIDENCE	No	95	Cap on AI-reported confidence scores
LIBTRACK_FETCH_DEBUG_MODE	No	False	Log raw API responses
LIBTRACK_API_RATE_LIMIT_SECONDS	No	1.5	Delay between registry API calls
LIBTRACK_DEDUP_WINDOW_HOURS	No	24	Hours before re-sending the same notification

*Required only when LIBTRACK_ENABLE_EMAIL_NOTIFICATIONS=True

---

Running Tests

pytest --cov=tracker

---

Contributing

1. Fork the repo and create a feature branch: git checkout -b feat/your-feature
2. Make your changes and add tests where relevant
3. Run pytest to confirm nothing is broken
4. Open a Pull Request with a clear description

---

Author

Built by Raghav Desai