CoverMe Complete User Guide

What happens behind the scenes:

• CoverMe automatically extracts your skills using AI
• Organizes them into 6 categories: Technical Skills, Frameworks, Domains, Industries, Key Strengths, Role Types
• Calculates your experience level from your resume
• All data stays on your computer - nothing is sent to external servers

Technical Skills Category:

Programming languages (JavaScript, Python, Java), technical tools (Docker, Kubernetes, Git). Add new skills using the input field and "Add" button.

Frameworks Category:

Web frameworks (React, Angular, Vue), backend frameworks (Django, Spring, Express), testing frameworks (Jest, Cypress, Selenium).

Domains Category:

Industry expertise (Healthcare, Finance, E-commerce), specialized knowledge areas.

Industries Category:

Sector experience (Technology, Healthcare, Manufacturing), market knowledge.

Key Strengths Category:

Soft skills (Leadership, Communication, Problem Solving), professional abilities.

Role Types Category:

Position types (Frontend Developer, Data Scientist, Product Manager), career focuses.

Method 1: Right-Click Context Menu (Universal)

Works on: Every website on the internet

How to use:

1. Navigate to any job posting on any website
2. Right-click anywhere on the page
3. Choose from three options:

Method 2: Manual Mode (Classic Workflow)

Best for: Detailed analysis when you want full control

Method 3: Live Mode (Automatic Monitoring)

Best for: Browsing multiple jobs quickly

How to activate:

1. Click the CoverMe extension icon
2. Toggle "Live Mode" to ON
3. Browse jobs normally

What happens automatically:

• Extension icon changes color as you navigate: 🟢 Green: 65%+ match (good fit) 🟠 Orange: 45-64% match (okay fit) 🔴 Red: <45% match (poor fit)
• Triggers on all recognized job sites
• Works across tabs - switch between job postings and see instant updates
• Content scripts auto-inject when needed
• Automatic error recovery if extension context breaks

Match Score Calculation:

• Green (65%+): Strong match - apply with confidence
• Orange (45-64%): Moderate match - highlight relevant experience
• Red (<45%): Weak match - consider if it's worth applying

Score Components:

• Job requirements fulfillment (75% weight)
• Skill relevance and strength (25% weight)
• Experience level penalties/bonuses

🎯 Well-Matched

Your experience aligns with job requirements. Sweet spot for applications.

⬆️ Overqualified

You have 5+ years more experience than required. Score penalty: -10% to -30% depending on gap. Consider: salary expectations, growth opportunities.

⬇️ Underqualified

You have 1+ years less experience than required. Score penalty: -8% to -25% depending on gap. Consider: emphasizing relevant projects, learning opportunities.

✅ Skills You Have That Job Requires

• Your matching skills found in the job posting
• Sorted by relevance and match confidence
• Click [+] to add skills not in your profile
• Green highlighting shows these on the job page

❌ Skills Job Requires That You Don't Have

• Critical missing skills identified by AI
• Prioritized by importance to the role
• Click [+] to add to your learning list
• Red highlighting shows these on the job page

🤔 Unknown Skills Found in Job

• Skills discovered by pattern matching
• Not in your current profile
• Confidence percentage (35-80%) based on extraction method
• [+] to add, [×] to ignore permanently

👤 Your Added Skills

• Skills you manually added during analysis
• Removable with [-] buttons
• Tracked across all job analyses
• Auto-categorized by AI during Deep Research

Multiple removal options:

1. [-] buttons: Remove individual skills you added
2. Options page: Bulk removal from keyword management interface
3. Ignored skills: Skills you [×] ignore are tracked and won't appear again

Happens automatically after Deep Research:

• Skills you add manually start in "Technical Skills"
• AI automatically moves them to proper categories:
  • React → Frameworks
  • Healthcare → Industries
  • Leadership → Key Strengths
• 👤 badges are removed once AI categorizes (now "officially organized")
• Process is non-critical - Deep Research succeeds even if categorization fails

5 intelligent extraction patterns:

1. Technology Names: Capitalized tech terms (React, Docker, AWS)
2. Experience Context: Skills mentioned with experience ("5 years of Python")
3. Framework Patterns: Common patterns ("framework", "library", "tool")
4. Capitalized Terms: Proper nouns that might be skills
5. Acronyms: Technical abbreviations (API, REST, CI/CD)

Smart filtering removes:

• Common words ("and", "the", "with")
• Generic terms ("experience", "knowledge", "skills")
• Skills already in your profile
• Previously ignored skills

Each job gets its own persistent chat thread:

• Identified by: Job Title + Company + URL
• Chat history persists across browser sessions
• Resume when you return to the same job posting
• Maximum 50 threads stored, oldest auto-deleted

🎯 Skills Match Tile

• Matches the format of keyword matching for easy comparison
• Shows your matching skills, missing skills, experience analysis
• Professional formatting with confidence scores

💼 Job Fit Assessment Tile

• Overall compatibility analysis
• Strengths you bring to the role
• Areas where you might need to demonstrate extra value
• Specific advice for this position

🎤 Interview Preparation Tile

• Questions likely to be asked based on job requirements
• Talking points for your experience
• How to address potential concerns
• Key things to research about the company/role

💡 Strategic Advice Tile

• Career guidance specific to this job and your background
• Whether you should apply and why
• Salary negotiation considerations
• Long-term career impact

🟢 Light Green Highlighting: Your Matching Skills

• Skills you have that the job requires
• Instantly shows your strengths on the job page
• Smooth gradient background with subtle animations
• High contrast for accessibility

🟣 Purple Highlighting: Experience Requirements

• Phrases like "5+ years experience", "minimum 3 years"
• Helps you quickly assess experience fit
• Distinguishes between hard requirements and preferences

🔴 Light Red Highlighting: Missing Critical Skills

• Skills the job requires that you don't have
• Prioritized by importance to the role
• Helps identify learning opportunities

Robust highlighting engine:

• Single-pass processing: Collects all matches first, then applies highlighting
• Overlap prevention: Smart conflict resolution when text matches multiple patterns
• Position-based sorting: Right-to-left replacement prevents DOM position shifts
• Ancestor protection: Won't highlight inside already-highlighted elements
• Performance optimized: Minimal impact on page loading and browsing

Cross-platform compatibility:

• Works on LinkedIn's dynamic content loading
• Handles Indeed's iframe structure
• Compatible with Glassdoor's lazy loading
• Adapts to company career page variations

Dashboard displays:

• Daily Usage: Current requests out of 250 daily limit
• Rate Limiting: Requests per minute (15 RPM max)
• Progress Bars: Color-coded (green → yellow → red as limits approach)
• Reset Timer: Shows when quota resets (midnight Pacific Time)

Automatic alerts:

• 80% Daily Limit: "You've used 200/250 daily requests"
• 90% Rate Limit: "You're approaching the rate limit (14/15 requests per minute)"
• Quota Exceeded: "Daily limit reached. Resets at midnight PT"

Tracked API calls:

• Ask AI Analysis: ~4,096 tokens per request
• Deep Research: ~6,144 tokens per request
• Keyword Extraction: ~2,048 tokens per request (one-time during setup)
• Skill Categorization: ~1,024 tokens per request (automatic during Deep Research)

Real-time statistics show exactly what's stored:

• Keywords: Count by category, total storage size
• Chat History: Number of threads, total messages, storage size
• Usage Analytics: Days tracked, total API calls made
• Job Analysis: Cached results, interaction history

Two export formats:

JSON Export (Developer-Friendly):

{
  "userProfile": {
    "keywords": {
      "technicalSkills": ["JavaScript", "Python"],
      "frameworks": ["React", "Django"]
    },
    "experienceYears": 5,
    "seniorityLevel": "Mid"
  },
  "chatHistory": [
    {
      "jobId": "senior-developer-techcorp",
      "messages": [...],
      "lastUpdate": "2025-01-18T10:30:00Z"
    }
  ],
  "usageAnalytics": {
    "dailyUsage": 150,
    "weeklyStats": [...]
  }
}

CSV Export (Spreadsheet-Friendly):

• Keywords sheet with categories
• Chat history with timestamps
• Usage analytics with dates
• Job interactions with outcomes

Local storage only:

• Everything stored in chrome.storage.local
• Encrypted by Chrome automatically
• No external servers except Google AI for analysis
• Google AI doesn't store your data permanently

Automatic cleanup:

• Chat threads auto-delete after 30 days
• Usage analytics auto-clean after 30 days
• User can override with "Keep data longer" setting

Adaptive recognition:

• When you use context menu on unrecognized sites, CoverMe asks: "Is this actually a job posting page?"
• If you confirm "Yes", it learns the URL pattern
• Future visits to similar pages on that site are automatically recognized
• Patterns stored: domain + first 1-2 path segments

Robust background system:

• Health check protocol: PING/PONG system ensures content scripts are responsive
• Auto-injection: Automatically injects content scripts when needed
• Context recovery: Handles extension context invalidation gracefully
• Cross-tab communication: Live mode state synced across all tabs

Never-fail philosophy:

• Local analysis: Always succeeds through multiple fallback strategies
• Content extraction: Multiple selectors and heuristic detection
• Extension recovery: Automatic page refresh for broken contexts (once per page)
• User guidance: Clear error messages with actionable next steps

Efficient processing:

• Background processing: Analysis happens without blocking UI
• Intelligent caching: Avoid re-analyzing the same job posting
• Minimal DOM impact: Highlighting adds <0.1s to page load
• Smart rate limiting: Queues requests to stay within Google's limits