Job Tracker 🚀

A comprehensive AI-powered Blazor application with browser extensions that automatically extract job listings from multiple job sites, analyze them with AI, and help you land your dream job faster.

🌟 What's New - AI Job Assistant

JobTracker now includes a complete AI-powered job search assistant! Leverage OpenAI GPT or Anthropic Claude to:

📊 Auto-analyze job descriptions - Extract key requirements, responsibilities, and skills
📝 Generate cover letters - Personalized suggestions based on your profile
🎯 Identify skill gaps - See what you have vs what's needed (with match %)
💼 Prep for interviews - Get likely technical and behavioral questions
💰 Negotiate salaries - Market rates, strategies, and scripts
📄 Optimize resumes - ATS keywords and tailored summaries
🔮 Predict success - AI estimates your application success probability
🔍 Find similar jobs - Discover related opportunities automatically

Choose your AI provider: OpenAI (GPT-3.5, GPT-4) or Anthropic Claude (3.5 Sonnet, Opus, etc.)

Supported Job Sites

Site	Extension Folder	Console Prefix	Status
LinkedIn	`BrowserExtensions/LinkedInExtension/`	`LJE`	Full support
Indeed	`BrowserExtensions/IndeedExtension/`	`IND`	Full support
S1Jobs	`BrowserExtensions/S1JobsExtension/`	`S1J`	Full support
Welcome to the Jungle	`BrowserExtensions/WTTJExtension/`	`WTTJ`	Full support
EnergyJobSearch	`BrowserExtensions/EnergyJobSearchExtension/`	`EJS`	Full support
Any site (Universal)	`BrowserExtensions/GenericExtension/`	`GEN`	Fallback support

Tip: Use the site-specific extensions (LinkedIn, Indeed, etc.) for the best experience on those sites — they have tailored extraction, auto-fetch, and crawl features. The Universal Extractor is a fallback for any other job site (Glassdoor, Monster, ZipRecruiter, etc.). If both are installed on the same site, the site-specific extension takes priority automatically.

Features

🤖 AI Job Assistant (NEW!)

Dual Provider Support - Choose between OpenAI or Anthropic Claude
Job Analysis - AI extracts summary, responsibilities, required skills, qualifications, and nice-to-have skills
Cover Letter Generation - Personalized opening, selling points, and closing based on your profile
Skill Gap Analysis - Visual comparison of your skills vs job requirements with match percentage
Similar Jobs - AI-powered recommendations based on skills, company, job type, and more
Interview Preparation - Generate 5 technical questions, 5 behavioral questions, and questions to ask them
Salary Negotiation - Market rate estimates, negotiation strategies, value points, and scripts
Resume Optimization - ATS keywords, skills to highlight, and tailored professional summary
Success Prediction - AI predicts application success probability with strength/risk factors
One-Click Analysis - "Analyze with AI" button on every job card
Persistent Results - All AI insights saved to job records for instant access
Model Selection - Choose from GPT-3.5 Turbo, GPT-4, GPT-4o, Claude 3.5 Sonnet, Claude 3 Opus, etc.
Cost Optimization - Switch models based on your budget and quality needs

Job Extraction

Multi-site Support - Extract jobs from LinkedIn, Indeed, S1Jobs, Welcome to the Jungle, EnergyJobSearch, and any site via the Universal Extractor
Automatic Extraction - Jobs are extracted automatically as you browse
Description Capture - Full job descriptions are captured when you view individual jobs
In-Page Description Fetch - Automatically clicks through job cards on list/collection pages to capture all descriptions
Auto-Fetch Descriptions - Automatically navigate through jobs to capture missing descriptions
Duplicate Detection - Prevents duplicate entries based on URL normalization
Source Tracking - Each job is tagged with its source site

Job Management

Dashboard - View all jobs with statistics (total, interested, applied, etc.)
Tabbed Interface - Browse, Possible, Applied, Pipeline (Kanban), Archived, and Unsuitable tabs
Pinned Jobs - Pin important jobs to the top of any tab
Filtering - Filter by search, location, job type, remote, salary, interest, source, skills, and date
Search Highlighting - Search terms highlighted in job title, company, location, and description preview
Notes & Cover Letter Search - Search finds matches in job titles, descriptions, companies, notes, and cover letters
Title-Only Search - Search specifically in job titles
Salary Range Filter - Parse salary strings into min/max values and filter by target salary
Skill Filter - Filter by skill with autocomplete from existing job skills
Saved Filter Presets - Save and load named filter configurations
Group By - Group job cards by company, title, location, job type, date posted/added, source, or salary band with collapsible accordion sections and expand/collapse all controls
Persisted View State - Filter, sort, group, and tab settings are automatically saved and restored between sessions
Bulk Actions - Mark all filtered jobs as Possible or Unsuitable with count feedback
Change Tracking - Track job listing changes over time with visual indicators:
- "Updated" Badge - Orange badge appears when job details change (description, salary, location, etc.)
- Change Timeline - View detailed change history with "View Changes" button
- Auto-Tracking - Changes are automatically detected when descriptions are fetched or updated
- Change Impact Levels - Major (title, company, salary), Moderate (location, description), and Minor changes
- First Description Tracking - Tracks when a job description is first added after initial import

Application Tracking

Interest Status - Mark jobs as Interested, Not Interested, or Not Rated
Suitability Status - Mark jobs as Possible, Unsuitable, or Not Checked
ML-Based Job Scoring - Automatic 0-100 scoring based on your preferences and past behavior (similar to JobSync)
- Skills match weighting
- Salary range matching
- Remote preference
- Location preferences
- Keyword matching (must-have and avoid)
- Company preferences
- Learning from your past application patterns
- Configurable weights for each factor
- Visual score badges on job cards (color-coded by score)
- Progress indicator during recalculation
- Skips unsuitable jobs during recalculation
Score-Based Rules - Create rules using ML scores (e.g., "Auto-mark jobs with score ≥80 as Interested")
Applied Tracking - Track which jobs you've applied to
Application Stages - Track progress: Applied → No Reply → Pending → Tech Test → Interview → Offer (or Ghosted/Rejected)
Stage History - View timeline of stage changes for each application
Pipeline Kanban Board - Drag-and-drop kanban board view of all applied jobs grouped by stage
Stale Application Alerts - Visual badges on pipeline cards when applications have had no stage change beyond a configurable threshold
Follow-up Reminders - Set per-job follow-up dates with due indicators and a count badge on the Pipeline tab
Cover Letter Templates - Reusable templates with {Company}, {Title}, {Location} placeholders that auto-fill when applied to a job
Auto-Archive - Automatically archive rejected/ghosted jobs after a configurable number of days
Auto-Delete - Automatically delete old jobs after configurable days:
- Delete unsuitable jobs after X days (default: 90)
- Delete rejected jobs after X days (default: 60)
- Delete ghosted jobs after X days (default: 60)
- Set to 0 to disable auto-deletion for that type

Application Stats Dashboard

Conversion Funnel - Visual funnel showing progression from Applied through to Offer/Rejected
Pipeline Health - Average days to reply, average days to interview, stale count, follow-ups due
Success & Ghosted Rates - Percentage breakdowns of application outcomes
Weekly Activity Charts - Jobs added and applications submitted over the last 12 weeks

Rules Engine

Auto-classification Rules - Automatically set interest, suitability, and remote status based on configurable rules
Rule Management - Create and manage rules from the Rules page (/rules)
Score-Based Rules - Use ML suitability scores in rules with numeric comparisons (>, >=, <, <=, =)
- Example: Auto-mark jobs with score ≥80 as "Interested"
- Example: Auto-reject jobs with score <30 as "Unsuitable"
- Combine score conditions with other criteria (AND/OR logic)
Compound Conditions - Create complex rules with multiple conditions
Priority System - Rules execute in priority order (higher priority first)
Rule Statistics - Track how many times each rule has triggered

ML Job Scoring (Similar to JobSync)

Automatic Scoring - Jobs are automatically scored 0-100 based on your preferences and behavior
Multi-Factor Analysis:
- Skills Match (25 points) - Matches against your preferred skills
- Salary Match (20 points) - Scores based on your desired salary range
- Remote Preference (15 points) - Preference for remote vs on-site
- Location Match (10 points) - Matches against preferred locations
- Keyword Matching (15 points) - Must-have and avoid keywords
- Company Preferences (10 points) - Preferred and avoided companies
- Behavioral Learning (15 points) - Learns from jobs you marked as interesting or applied to
Configurable Weights - Adjust the importance of each factor (0-1 scale)
Visual Score Badges - Color-coded badges on job cards:
- 🟢 Green (80-100): Excellent match
- 🔵 Blue (60-79): Good match
- 🟡 Yellow (40-59): Fair match
- ⚫ Grey (20-39): Poor match
- 🔴 Red (0-19): Very poor match
Smart Recalculation - Recalculate all scores with live progress indicator
- Automatically skips unsuitable jobs
- Shows progress bar and percentage
- Runs in background without blocking UI
Min Score Filter - Hide jobs below a certain score threshold
Integration with Rules - Use scores in automation rules for smart job triage

Background Crawling

Server-side Crawling - Crawl job sites directly from the server without browser extensions
Supported Crawl Sites - LinkedIn, S1Jobs, Welcome to the Jungle, and EnergyJobSearch
Scheduled Crawling - Automatic crawling via Hangfire (SQL Server mode) or LocalBackgroundService
Ghosted Detection - Automatically marks old applications with no reply as Ghosted (configurable threshold)
No Reply Detection - Automatically marks applied jobs as No Reply after configurable days
Availability Checks - Background checks for job listing availability
Auto-Archive Job - Background job to archive rejected/ghosted jobs after configurable days
Job Cleanup Job - Automatically deletes old unsuitable, rejected, and ghosted jobs based on configurable thresholds
Email Notification Job - Daily digest email for follow-ups due and stale applications
IMAP Retry Logic - Email fetch with exponential backoff (3 attempts with 2s, 4s delays)
Data Corruption Protection - Automatic backup of corrupted JSON data files before returning defaults
API Input Validation - Length limits on URLs (2000), locations (500), salaries (500), and descriptions (100K)
Configurable Thresholds - No Reply, Ghosted, Stale, and Auto-Delete days are all configurable per-user in Settings

Authentication (SQL Server mode)

User Accounts - Login with cookie-based authentication and API key support
Two-Factor Authentication - Optional 2FA setup and verification
Password Recovery - Forgot password and reset password flows

Appearance & Accessibility

Dark Mode - Toggle dark/light theme with one click; persists across sessions via localStorage
Keyword Highlighting - Configurable keyword highlighting in job descriptions
Search Highlighting - Matching search terms highlighted in yellow across job cards
ARIA Labels - Proper accessibility labels on all filter controls, tabs, and interactive elements for screen reader support

Notifications

Email Notifications - Daily digest emails for follow-ups due and stale applications
Per-User SMTP Configuration - Each user configures their own SMTP settings in Settings page
No Configuration File Needed - SMTP settings are managed entirely through the UI (no appsettings.json required)
Password Reset Emails - Uses recipient's SMTP settings for password reset notifications

Sharing & Export

WhatsApp Sharing - Share job details directly to WhatsApp
CSV Export - Export applied jobs to CSV from the Jobs page
History Export - Export full or filtered history to CSV
Direct Links - Open original job postings with one click

Crawl Pages (NEW)

Custom crawl URL list (/crawl-pages) - Add additional search/listing URLs to be crawled alongside your main site URLs.
Per-page delay - Set how many seconds to wait after each crawl page (default: 90s) to be polite to job sites.
Optional {n} substitution - Define a start/end/increment range to expand a single URL template into multiple pages.
- Example: ...&page={n} with Start=1, End=5, Increment=1.
Run in Browser - Opens each enabled crawl page in a new tab and automatically closes it after its delay (requires allowing popups for https://localhost:7046).
Reorder pages - Use up/down arrows to control crawl order.

Portable / Pre-built (No SDK Required)

Pre-built self-contained executables are available in the Portable/ folder. No .NET SDK or runtime install is needed.

Files	Platform
`Portable/JobTracker-x86.zip.001`, `.002`	Windows 32-bit
`Portable/JobTracker-x64.zip.001`, `.002`	Windows 64-bit

The archives are split into 80MB parts for easier distribution. You need 7-Zip to extract them.

Quick Start

Download all .zip.001, .zip.002, etc. parts for your platform into the same folder
Right-click the .zip.001 file and choose 7-Zip > Extract Here (or run 7z x JobTracker-x86.zip.001)
Edit appsettings.json if needed (defaults work out of the box)
Run JobTracker.exe
Open https://localhost:7046 in your browser
Accept the self-signed certificate warning on first visit

The app creates a Data/ folder alongside the executable for storing jobs, settings, and history. To update, just replace the files (keep your Data/ folder and appsettings.json).

Note: These zips are rebuilt automatically on every Release build. If you're building from source, run dotnet build -c Release and the split zips appear in Portable/.

Prerequisites

.NET 10 SDK or later (only needed if building from source)
Google Chrome or Microsoft Edge browser
Accounts on job sites you want to use

Installation (From Source)

1. Set Up the Blazor Application

# Navigate to the project directory
cd JobTracker

# Restore dependencies
dotnet restore

# Run the application
dotnet run

The app will start at https://localhost:7046.

2. Install Browser Extensions

Each job site has its own extension. Install the ones you need:

Open Chrome/Edge and navigate to chrome://extensions/ or edge://extensions/
Enable Developer mode (toggle in the top-right corner)
Click Load unpacked
Select the extension folder:
- BrowserExtensions/LinkedInExtension/ - LinkedIn
- BrowserExtensions/IndeedExtension/ - Indeed
- BrowserExtensions/S1JobsExtension/ - S1Jobs
- BrowserExtensions/WTTJExtension/ - Welcome to the Jungle
- BrowserExtensions/EnergyJobSearchExtension/ - EnergyJobSearch
- BrowserExtensions/GenericExtension/ - Universal Extractor (any job site)
Repeat for each extension you want to install

Universal Extractor: Opt-in per site — on each new domain a small prompt asks you to activate. Once activated, it uses three extraction strategies: schema.org JSON-LD metadata, DOM job link detection, and heuristic fallback. It automatically defers to site-specific extensions when both are installed.

Usage

Extracting Jobs

Start the Blazor app - Make sure it's running (dotnet run)
Navigate to a job site - Go to LinkedIn Jobs, Indeed, S1Jobs, WTTJ, EnergyJobSearch, or any job site (with the Universal Extractor)
Browse jobs - The extension automatically extracts job listings as you scroll
View job details - Click on individual jobs to capture full descriptions
Check the indicator - A colored pill in the bottom-right shows extraction status

Auto-Fetch Descriptions

Each extension can automatically navigate through jobs to capture missing descriptions:

Via Console - Open DevTools (F12) and type:
```
// LinkedIn
LJE.autoFetch(3)    // 3 second delay between jobs

// Indeed
IND.autoFetch(3)

// S1Jobs
S1J.autoFetch(3)

// WTTJ
WTTJ.autoFetch(3)

// EnergyJobSearch
EJS.autoFetch(3)
```
Note: The Universal Extractor (GEN) does not support auto-fetch as it has no site-specific card selectors. Use the Extract button in the popup or GEN.extract() in the console instead.
Via Extension Popup - Click the extension icon and press "Auto-Fetch Descriptions"
Stop Auto-Fetch - Click the Stop button in the UI or run:
```
LJE.stopAutoFetch()  // or IND, S1J, WTTJ, EJS
```

Extension Console Commands

Each extension exposes commands in the browser console:

Command	Description
`XXX.extract()`	Manually trigger job extraction
`XXX.debug()`	Show debug info and page structure
`XXX.test()`	Test connection to the Blazor app
`XXX.status()`	Show extraction statistics
`XXX.getDesc()`	Get current job's description
`XXX.updateDesc()`	Send current description to server
`XXX.fetchMissing()`	Check for jobs needing descriptions
`XXX.autoFetch(n)`	Start auto-fetch with n second delay
`XXX.stopAutoFetch()`	Stop auto-fetching
`XXX.autoFetchStatus()`	Check auto-fetch progress
`XXX.help()`	Show all available commands

Replace XXX with: LJE (LinkedIn), IND (Indeed), S1J (S1Jobs), WTTJ (Welcome to the Jungle), EJS (EnergyJobSearch), or GEN (Universal Extractor)

Universal Extractor (GEN) commands: GEN.extract(), GEN.debug(), GEN.test(), GEN.status(), and GEN.help() are available. Auto-fetch commands are not supported by the Universal Extractor.

Managing Jobs in the Web App

Navigate to https://localhost:7046 to access the dashboard.

Pages

Page	Path	Description
Jobs Dashboard	`/` or `/jobs`	Main job listing and management with AI analysis
Add Job	`/jobs/add`	Manually add a job
Dashboard	`/dashboard`	Application stats, conversion funnel, and weekly activity
Rules	`/rules`	Manage auto-classification rules (including ML score-based rules)
Settings	`/settings`	Configure user profile, 2FA, browser extension API key, job site URLs, AI Assistant, ML scoring preferences, highlight keywords, pipeline thresholds, auto-archive/delete settings, SMTP/email settings, and cover letter templates
History	`/history`	View audit log of changes with CSV export
Background Jobs	`/background-jobs`	Monitor background crawl jobs (LocalMode only)
Extension Install	`/extension-install`	Extension installation guide
About	`/about`	Application information

Tabs

Browse - New jobs that haven't been categorized
Possible - Jobs marked as potentially suitable
Applied - Jobs you've applied to (list view)
Pipeline - Kanban board of applied jobs grouped by stage (drag-and-drop)
Archived - Jobs that have been archived (manually or auto-archived)
Unsuitable - Jobs that don't match your criteria

Job Cards

Click the card header to open the job on the original site
Use thumbs up/down buttons to set interest
Use Possible/Unsuitable buttons to categorize
Click "Applied" to track applications
Use the stage dropdown to update application progress
Pin/unpin jobs to keep them at the top of any tab
Archive/unarchive jobs
Click WhatsApp icon to share
Click "View Full Description" for details
Click "Fetch Details" to refresh job information from the job page (description, salary/location if available; some sites may limit what can be fetched server-side)

"Fetch Details" also re-applies your rules after updating the job so auto-classification stays consistent.

Contacts

JobTracker includes a built-in contact manager so you can track recruiters, hiring managers, and your interactions.

Where contacts live

Contacts page (/contacts): browse/search all contacts and view interaction history.
Per-job Contacts tab: each job has a Contacts tab in the job details modal where you can add/edit/remove contacts and log interactions (Email/Phone/LinkedIn/etc.).

Capturing contacts from job sites

Contacts can be associated with jobs in two ways:

Manual entry (always works): open a job → Contacts tab → Add Contact.
Automatic capture (via browser extensions where supported): extensions can send recruiter/contact data along with description updates.

When an extension updates a job via PUT /api/jobs/description, it may include a Contacts array. The server will merge/dedupe contacts by name and link them to the matching job.

Filtering Options

Filter	Description
Search	Search in title, description, company, notes, and cover letter
Title Only	Search only in job titles
Location	Filter by job location
Job Type	Full-time, Part-time, Contract, etc.
Remote	Remote only or On-site only
Source	LinkedIn, Indeed, S1Jobs, WTTJ, EnergyJobSearch, or other (via Universal Extractor)
Interest	Interested, Not Interested, or Not Rated
Salary Info	Has salary or No salary listed
Salary Contains	Search within salary text (e.g., "100K")
Salary Target	Filter jobs whose parsed salary range covers the target value
Skill	Filter by skill with autocomplete
Stage	Filter applied jobs by application stage
Date Range	Filter by date posted
Sort By	Sort by date added, date posted, title, company, or salary
Group By	Group by company, title, location, job type, date posted/added, source, or salary band

API Endpoints

Method	Endpoint	Description
GET	`/api/jobs`	Get all jobs
POST	`/api/jobs`	Add a new job (with duplicate detection)
GET	`/api/jobs/count`	Get total job count
GET	`/api/jobs/stats`	Get job statistics
GET	`/api/jobs/exists?url=...`	Check if job URL exists
PUT	`/api/jobs/{id}/interest`	Update interest status
PUT	`/api/jobs/description`	Update job description by URL
GET	`/api/jobs/needing-descriptions`	Get jobs missing descriptions
POST	`/api/jobs/cleanup`	Clean up job titles
POST	`/api/jobs/remove-duplicates`	Remove duplicate jobs
POST	`/api/jobs/fix-sources`	Fix unknown sources from URLs
DELETE	`/api/jobs/{id}`	Delete a job
DELETE	`/api/jobs/clear`	Delete all jobs

Data Storage

The app supports two storage backends, controlled by the StorageProvider setting in appsettings.json.

JSON (default)

Jobs, history, and settings are stored as JSON files in the Data/ directory:

Data/jobs.json - Job listings
Data/history.json - Audit/history log
Data/settings.json - App settings and rules

No configuration needed - this is the default when StorageProvider is "Json" or not set.

SQL Server

To switch to SQL Server, update appsettings.json:

{
  "StorageProvider": "SqlServer",
  "ConnectionStrings": {
    "JobSearchDb": "Server=myserver;Database=JobSearch;User Id=myuser;Password=mypassword;TrustServerCertificate=True"
  }
}

On first startup with SQL Server:

The database and tables are created automatically
If the JSON data files exist and the database is empty, all data is imported automatically
No manual migration steps are needed

To switch back to JSON, set "StorageProvider": "Json". The JSON files are never modified while running in SQL Server mode, so they remain as they were.

Tables (SQL Server)

Table	Contents
`Users`	User accounts and authentication
`JobListings`	All tracked jobs (with parsed salary, follow-up dates)
`HistoryEntries`	Audit log of changes
`JobRules`	Auto-classification rules
`AppSettings`	Per-user config (site URLs, rule settings, pipeline thresholds, dark mode, email, auto-archive, cover letter templates)

Project Structure

JobTracker/
├── Components/
│   ├── Pages/
│   │   ├── Jobs.razor              # Main job dashboard (browse, filter, kanban pipeline)
│   │   ├── AddJob.razor            # Manual job entry
│   │   ├── Dashboard.razor         # Application stats and conversion funnel
│   │   ├── Rules.razor             # Auto-classification rules
│   │   ├── Settings.razor          # App settings
│   │   ├── History.razor           # Change audit log with CSV export
│   │   ├── BackgroundJobs.razor    # Background job monitoring
│   │   └── ExtensionInstall.razor  # Extension setup guide
│   └── Layout/
│       └── NavMenu.razor           # Navigation menu
├── Models/
│   ├── JobListing.cs               # Job data model (with salary parsing, follow-up dates)
│   ├── AppSettings.cs              # Settings, site URLs, pipeline thresholds, filter presets, cover letter templates
│   ├── JobHistory.cs               # History entry model
│   └── JobRule.cs                  # Rule model
├── Services/
│   ├── JobListingService.cs        # Job storage, filtering, pipeline stats
│   ├── SalaryParser.cs             # Salary string parser (min/max extraction)
│   ├── JobScoringService.cs        # ML-based job scoring engine (0-100 scoring)
│   ├── AIJobAssistantService.cs    # 🤖 AI-powered job analysis (OpenAI & Claude)
│   ├── JobCrawlService.cs          # Server-side site crawling
│   ├── JobCrawlJob.cs              # Scheduled crawl job
│   ├── GhostedCheckJob.cs          # Auto-ghosted detection (configurable threshold)
│   ├── NoReplyCheckJob.cs          # Auto no-reply detection (configurable threshold)
│   ├── AvailabilityCheckJob.cs     # Job availability checks
│   ├── AutoArchiveJob.cs           # Auto-archive rejected/ghosted jobs
│   ├── JobCleanupJob.cs            # Auto-delete old unsuitable/rejected/ghosted jobs
│   ├── EmailNotificationJob.cs     # Daily digest email notifications
│   ├── EmailCheckJob.cs             # IMAP email check with retry logic
│   ├── EmailReplyMatcher.cs         # Match emails to jobs and detect application stages
│   ├── EmailService.cs             # SMTP email sending
│   └── LocalBackgroundService.cs   # Background task runner
├── Data/
│   └── jobs.json                   # Persistent job storage (JSON mode)
├── BrowserExtensions/
│   ├── LinkedInExtension/
│   │   ├── manifest.json
│   │   ├── content.js
│   │   ├── popup.html
│   │   └── popup.js
│   ├── IndeedExtension/
│   │   ├── manifest.json
│   │   ├── content.js
│   │   ├── popup.html
│   │   └── popup.js
│   ├── S1JobsExtension/
│   │   ├── manifest.json
│   │   ├── content.js
│   │   ├── popup.html
│   │   └── popup.js
│   ├── WTTJExtension/
│   │   ├── manifest.json
│   │   ├── content.js
│   │   ├── popup.html
│   │   └── popup.js
│   ├── EnergyJobSearchExtension/
│   │   ├── manifest.json
│   │   ├── content.js
│   │   ├── popup.html
│   │   └── popup.js
│   └── GenericExtension/
│       ├── manifest.json
│       ├── content.js
│       ├── background.js
│       ├── popup.html
│       └── popup.js
└── Program.cs                      # App configuration and API endpoints

Configuration

🤖 AI Job Assistant Setup

1. Get an API Key

Option A: OpenAI

Visit OpenAI API Keys
Create a new API key (starts with sk-)
Cost: ~$0.01-0.05 per job analysis depending on model

Option B: Anthropic Claude

Visit Anthropic Console
Create a new API key (starts with sk-ant-)
Cost: ~$0.008-0.08 per job analysis depending on model

2. Configure in Settings

Navigate to Settings → AI Job Assistant section
Enable AI Assistant (check the box)
Select AI Provider: OpenAI or Claude
Enter your API Key
Choose Model:
- OpenAI: GPT-3.5 Turbo (fast/cheap), GPT-4 (accurate), GPT-4o (latest)
- Claude: Claude 3 Haiku (fast/cheap), Claude 3.5 Sonnet (recommended), Claude 3 Opus (most capable)
Configure Your Profile:
- Your Skills - Comma-separated (e.g., "C#, .NET, Azure, React, TypeScript")
- Experience Summary - Brief description of your experience
Enable AI Features:
- ✅ Auto-analyze new jobs
- ✅ Auto-generate cover letter suggestions
- ✅ Show skill gap analysis
- ✅ Show similar job recommendations
Save AI Settings

3. Using AI Features

On Job Cards:

Click 🤖 Analyze with AI button to analyze any job
Results appear in a beautiful modal with tabbed sections
All insights are automatically saved to the job

AI Analysis Modal:

📊 Summary - 2-3 sentence job overview
✅ Responsibilities - Key duties extracted
⚙️ Required Skills - Color-coded (green = you have it)
🎓 Qualifications - Education and experience needs
⭐ Nice-to-Have Skills - Optional bonus skills
📈 Skill Match - Percentage with progress bar

Advanced Features (from modal):

📋 Interview Prep - Get likely interview questions
💰 Salary Tips - Negotiation strategies and scripts
📄 Resume Tips - ATS keywords and optimization
📊 Success Prediction - AI-powered success probability
🔍 Similar Jobs - Find related opportunities
✉️ Cover Letter - Personalized writing suggestions

AI Provider Comparison

Feature	OpenAI	Claude
Best For	Code, structured output	Analysis, writing, reasoning
Context Window	8K-128K tokens	200K tokens
Pricing	Moderate	Competitive
Recommended Model	GPT-3.5 Turbo or GPT-4o	Claude 3.5 Sonnet
Cost per Analysis	$0.01-0.05	$0.008-0.08

Tip: Start with GPT-3.5 Turbo or Claude 3.5 Sonnet for best balance of quality and cost!

🎯 AI Features Showcase

📊 Job Analysis

Extract structured information from any job posting:

✅ Summary: "Senior .NET Developer role focusing on cloud architecture..."
✅ Responsibilities: ["Design scalable APIs", "Lead technical reviews"...]
✅ Required Skills: ["C#", ".NET Core", "Azure", "SQL Server"...]
✅ Qualifications: ["Bachelor's in CS", "5+ years .NET experience"...]
✅ Nice-to-Have: ["Docker", "Kubernetes", "React"...]

🎯 Skill Gap Analysis

Visual comparison with percentage match:

Match: 85% ✅

Skills You Have (7):
✅ C#  ✅ .NET  ✅ Azure  ✅ SQL Server  ✅ REST APIs  ✅ Git  ✅ Agile

Missing Required Skills (2):
❌ Kubernetes  ❌ Terraform

Bonus Skills You Have (2):
⭐ Docker  ⭐ React

📝 Cover Letter Assistant

AI generates personalized suggestions:

Opening: "As a passionate .NET architect with 8 years of cloud experience..."

Selling Points:
• Extensive Azure expertise aligns with your cloud-first approach
• Proven track record leading teams of 5+ developers
• Strong background in microservices and container orchestration

Closing: "I'm excited to bring my technical leadership to [Company]..."

💼 Interview Preparation

Get likely questions before the interview:

Technical Questions:
1. How would you design a microservices architecture on Azure?
2. Explain your approach to optimizing database queries in SQL Server
3. Describe a complex technical challenge you solved recently

Behavioral Questions:
1. Tell me about a time you had to mentor junior developers
2. How do you handle conflicting priorities in agile sprints?

Questions to Ask Them:
1. What's the team's approach to DevOps and CI/CD?
2. How does the company support professional development?

💰 Salary Negotiation

Market data and professional scripts:

Market Rate: £60,000-£75,000 for Senior .NET Developer in London

Strategies:
• Research shows market rate 15% above initial offer
• Emphasize your Azure certifications and leadership experience
• Focus on total compensation including benefits

Opening Statement:
"I'm very excited about this opportunity. Based on my research and 
the value I bring, I was hoping we could discuss £70,000..."

📄 Resume Optimization

ATS keywords and tailored content:

Keywords to Include:
"microservices", "Azure DevOps", "CI/CD", "Agile", "Scrum Master"

Skills to Highlight:
1. Azure Architecture (Solution Architect certified)
2. .NET Core 6/7 (8+ years)
3. Team Leadership (Led teams of 5+)

Tailored Summary:
"Senior .NET Developer with 8 years of cloud-native development, 
specializing in Azure microservices and team leadership..."

🔮 Success Prediction

AI-powered probability estimate:

Success Probability: 78% 📈

Strength Factors:
✅ Strong Azure experience matches requirements
✅ Leadership background aligns with senior role
✅ Technical skills cover 85% of requirements

Risk Factors:
⚠️ No Kubernetes experience (required skill)
⚠️ Company prefers finance sector background

Recommendation: Apply Now ✅
Tip: Highlight transferable skills to address Kubernetes gap

🔍 Similar Jobs

AI finds related opportunities:

Similar Jobs Found: 5

1. Cloud Solutions Architect @ TechCorp (92% match)
   Why: 8 shared skills, similar role, remote

2. Senior .NET Developer @ DataSys (87% match)
   Why: Same stack, both emphasize Azure, London-based

3. Technical Lead @ CloudFirst (82% match)
   Why: Leadership focus, cloud-native, remote

ML Job Scoring Setup

Navigate to Settings - Go to /settings or click Settings in the sidebar
Scroll to "Job Scoring / ML Preferences" section
Enable ML-based Job Scoring - Check the checkbox
Configure Scoring Weights (0 = disabled, 1 = full weight):
- Skills Match Weight
- Salary Match Weight
- Remote Preference Weight
- Location Weight
- Keywords Weight
- Company Preference Weight
- Learning from Behavior Weight
Set Your Preferences:
- Preferred Skills - Comma-separated (e.g., "C#, .NET, Azure, React")
- Salary Range - Minimum and maximum desired salary
- Remote Preference - Check if you prefer remote jobs
- Preferred Locations - Comma-separated locations
- Must-Have Keywords - Keywords that should be in job descriptions
- Avoid Keywords - Keywords that indicate jobs you want to skip
- Preferred Companies - Companies you'd like to work for
- Avoid Companies - Companies to skip
- Minimum Score to Show - Hide jobs below this threshold (0 = show all)
Save Scoring Preferences
Recalculate All Scores - Click to score all existing jobs (with live progress)
View Scores - Go to Jobs page to see score badges on each job

Score-Based Rules

Create automation rules using ML scores:

Go to Rules page
Click Add New Rule
Select Field: Suitability Score
Select Operator: >=, >, <=, <, or =
Enter Value: Score threshold (0-100)
Set Action: Set Interest, Suitability, or IsRemote
Example Rules:
- Score >= 80 → Set Interest = Interested
- Score < 30 → Set Suitability = Unsuitable
- Score >= 70 AND Location Contains "Remote" → Set Interest = Interested

Email Configuration

Navigate to Settings → Email / SMTP section
Configure SMTP Settings:
- SMTP Host (e.g., smtp.gmail.com)
- Port (e.g., 587)
- Username and Password
- From Email and From Name
Enable Email Notifications (optional)
Configure Notification Preferences:
- Enable daily email digest
- Follow-ups due notifications
- Stale applications notifications

Note: Each user manages their own SMTP settings. There is no shared SMTP configuration.

Auto-Delete Configuration

Navigate to Settings → Auto-Archive & Auto-Delete section
Configure Delete Thresholds:
- Delete Unsuitable - Days to keep unsuitable jobs (0 = don't delete)
- Delete Rejected - Days to keep rejected applications (0 = don't delete)
- Delete Ghosted - Days to keep ghosted applications (0 = don't delete)
Save Settings
Background job will automatically delete old jobs based on these thresholds

Troubleshooting

Extension not detecting jobs

Make sure you're on a job listing page
Open DevTools (F12) → Console tab
Look for [LJE], [IND], [S1J], [WTTJ], [EJS], or [GEN] messages
Run the debug() command to see page structure
Try the extract() command to manually trigger

Connection failed

Verify the Blazor app is running (dotnet run)
Visit https://localhost:7046 directly to accept any certificate warnings
Check the extension popup shows "Connected"

Jobs not appearing in the app

Check the Blazor app console for [API] messages
Verify the job was added (check for "Added:" in extension console)
Jobs might be filtered - try clearing all filters

Missing descriptions

Click on individual jobs to capture their descriptions

Use auto-fetch to automatically capture descriptions:

LJE.autoFetch(3)  // Replace LJE with appropriate prefix

Unknown source on jobs

Run the fix-sources endpoint to infer sources from URLs:

curl -X POST https://localhost:7046/api/jobs/fix-sources

Universal Extractor not activating

The Universal Extractor uses an opt-in model — it only runs on sites you explicitly activate. On each new domain, a small prompt appears in the bottom-right corner asking "Activate Job Tracker for this site?". Click Activate to enable extraction for that domain permanently. You can also manage activated sites from the extension popup.

Duplicate text in job titles

Run the cleanup endpoint:

curl -X POST https://localhost:7046/api/jobs/cleanup

Application Stages

When tracking applications, jobs progress through these stages:

Stage	Description
Applied	Initial application submitted
No Reply	No response received yet
Pending	Awaiting decision
Ghosted	No response after extended time
Rejected	Application was rejected
Tech Test	Technical assessment stage
Interview	Interview stage
Offer	Received job offer

Click the stage badge on a job card to cycle through stages, or use the dropdown to select a specific stage.

🛠️ Tech Stack

Backend

.NET 10 - Latest .NET framework
Blazor Server - Interactive server-side rendering
C# 14 - Modern C# language features
SQL Server - Optional enterprise database support
Hangfire - Background job processing (SQL Server mode)
Entity Framework Core - ORM for SQL Server

AI & ML

OpenAI GPT - GPT-3.5 Turbo, GPT-4, GPT-4o support
Anthropic Claude - Claude 3 Haiku, Sonnet, Opus support
Custom ML Scoring - Multi-factor job scoring algorithm
JSON Deserialization - Structured AI response parsing

Frontend

Blazor Components - Reusable UI components
Bootstrap 5 - Responsive design system
Bootstrap Icons - Icon library
Dark Mode - Theme switching with localStorage
Drag & Drop - Kanban board (SortableJS integration)

Browser Extensions

Chrome Extension API - Manifest V3
JavaScript - Content scripts and background workers
DOM Manipulation - Job extraction from various sites
Cross-Origin Messaging - Extension to app communication

Data Storage

JSON Files - Default lightweight storage
SQL Server - Enterprise-grade option
Entity Framework Migrations - Schema management
Dual Backend Architecture - Seamless switching

Authentication & Security

Cookie Authentication - Session management
API Key Support - Extension authentication
TOTP 2FA - Time-based one-time passwords
Password Hashing - Secure credential storage
SMTP Email - Password reset and notifications

Background Jobs

LocalBackgroundService - JSON mode background tasks
Hangfire - SQL Server mode job scheduling
Recurring Jobs - Crawling, email, cleanup
CRON Scheduling - Flexible job timing

📊 Stats & Metrics

9 AI Features - Comprehensive job search assistance
2 AI Providers - OpenAI and Anthropic Claude
8 Model Options - From fast/cheap to most capable
6 Job Sites - LinkedIn, Indeed, S1Jobs, WTTJ, EnergyJobSearch + Universal Extractor
7 Application Stages - Full pipeline tracking
ML Scoring - 7-factor intelligent job matching
Kanban Pipeline - Visual drag-and-drop board
Background Jobs - 7 automated tasks
Dual Storage - JSON or SQL Server
Browser Extensions - 5 dedicated extractors + 1 universal fallback
184 Unit Tests - EmailReplyMatcher, JsonStorageBackend, SalaryParser, JobScoring, JobRules, CompanyNormalization, JobTypeDetection

🎉 Why JobTracker?

✅ AI-Powered - Leverage GPT-4 or Claude to work smarter
✅ Comprehensive - Track every aspect of your job search
✅ Intelligent - ML scoring learns from your preferences
✅ Automated - Rules engine and background jobs do the work
✅ Visual - Kanban boards, charts, and color-coded indicators
✅ Flexible - Choose your AI provider, storage, and workflows
✅ Private - Your data stays on your server
✅ Extensible - Open architecture for customization

License

This project is for personal use to track job applications.

Contributing

Feel free to submit issues and enhancement requests!

Name		Name	Last commit message	Last commit date
Latest commit History 66 Commits
.claude		.claude
.github/workflows		.github/workflows
BrowserExtensions		BrowserExtensions
Components		Components
Data		Data
JobTracker.Tests		JobTracker.Tests
Migrations		Migrations
Models		Models
Portable		Portable
Properties		Properties
Services		Services
build		build
images		images
wwwroot		wwwroot
.gitattributes		.gitattributes
.gitignore		.gitignore
JobTracker.csproj		JobTracker.csproj
JobTracker.slnx		JobTracker.slnx
Program.cs		Program.cs
README.md		README.md
appsettings.Development.json		appsettings.Development.json
appsettings.json		appsettings.json
dotnet-tools.json		dotnet-tools.json

Folders and files

Latest commit

History

Repository files navigation

Job Tracker 🚀

🌟 What's New - AI Job Assistant

Supported Job Sites

Features

🤖 AI Job Assistant (NEW!)

Job Extraction

Job Management

Application Tracking

Application Stats Dashboard

Rules Engine

ML Job Scoring (Similar to JobSync)

Background Crawling

Authentication (SQL Server mode)

Appearance & Accessibility

Notifications

Sharing & Export

Crawl Pages (NEW)

Portable / Pre-built (No SDK Required)

Quick Start

Prerequisites

Installation (From Source)

1. Set Up the Blazor Application

2. Install Browser Extensions

Usage

Extracting Jobs

Auto-Fetch Descriptions

Extension Console Commands

Managing Jobs in the Web App

Pages

Tabs

Job Cards

Contacts

Where contacts live

Capturing contacts from job sites

Filtering Options

API Endpoints

Data Storage

JSON (default)

SQL Server

Tables (SQL Server)

Project Structure

Configuration

🤖 AI Job Assistant Setup

1. Get an API Key

2. Configure in Settings

3. Using AI Features

AI Provider Comparison

🎯 AI Features Showcase

📊 Job Analysis

🎯 Skill Gap Analysis

📝 Cover Letter Assistant

💼 Interview Preparation

💰 Salary Negotiation

📄 Resume Optimization

🔮 Success Prediction

🔍 Similar Jobs

ML Job Scoring Setup

Score-Based Rules

Email Configuration

Auto-Delete Configuration

Troubleshooting

Extension not detecting jobs

Connection failed

Jobs not appearing in the app

Missing descriptions

Unknown source on jobs

Universal Extractor not activating

Duplicate text in job titles

Application Stages

🛠️ Tech Stack

Backend

AI & ML

Frontend

Browser Extensions

Data Storage

Authentication & Security

Packages