Resume CLI System

A unified command-line interface for generating job-specific resumes from a single YAML source. Supports fast template-based generation and AI-powered customization with full integration into application tracking.

Features

• Single Source of Truth: Store all resume data in resume.yaml
• Fast Generation: Create any variant in <1 second using Jinja2 templates
• AI-Powered: Optional AI customization using Claude or OpenAI
• ATS Compatibility: Check resume optimization for automated screening systems
• Multiple Variants: 6 pre-configured variants (base, backend, ML/AI, fullstack, DevOps, leadership)
• Multiple Formats: Output to Markdown, LaTeX, or PDF
• Application Tracking: Built-in CSV tracking with analytics
• GitHub Integration: Sync projects automatically
• LinkedIn Integration: Import/export LinkedIn profile data (including full data archive with folder export)
• Schema Validation: Catch errors before generating

Installation

Prerequisites

• Python: 3.8, 3.9, 3.10, 3.11, or 3.12 (tested in CI)
• LaTeX (optional, for PDF generation): See PDF Generation section

From Source

git clone https://github.com/anchapin/resume-cli.git
cd resume-cli
pip install -e .

With AI Support

pip install -e ".[ai]"

This installs the required AI API packages:

• anthropic for Claude
• openai for GPT models
• python-dotenv for environment variable management

Quick Start

1. Set up your resume data

Option A: Start from scratch (recommended)

Copy the example resume file and customize it:

cp resume.example.yaml resume.yaml
# Edit resume.yaml with your information

Option B: Import from existing resume

If you have existing resume files:

resume-cli init --from-existing

This parses resumes/base_resume.txt and creates resume.yaml.

2. Generate a resume

# Generate base variant as Markdown
resume-cli generate -v v1.0.0-base -f md

# Generate backend variant as PDF
resume-cli generate -v v1.1.0-backend -f pdf

# List all available variants
resume-cli variants

3. Validate your data

resume-cli validate

Commands

generate

Generate a resume from template or AI.

resume-cli generate [OPTIONS]

Options:
  -v, --variant TEXT      Resume variant (default: v1.0.0-base)
  -f, --format CHOICE     Output format: md, tex, pdf (default: md)
  -o, --output PATH       Output file path
  --no-save              Print to stdout without saving
  --ai                   Use AI-powered generation (requires --job-desc)
  --job-desc PATH        Path to job description file (required for --ai)

Examples:

# Generate Markdown resume (template-based)
resume-cli generate -v v1.0.0-base -f md

# Generate PDF with custom output path
resume-cli generate -v v1.1.0-backend -f pdf -o my-resume.pdf

# AI-customized for specific job
resume-cli generate --ai --job-desc job-posting.txt

# Preview without saving
resume-cli generate -v v1.2.0-ml_ai --no-save

Note: The --ai flag requires --job-desc. Without --ai, generation uses fast templates (<1 second).

variants

List all available resume variants.

resume-cli variants

ats-check

Check ATS (Applicant Tracking System) compatibility score.

Analyzes resume against job description and provides actionable feedback for optimization.

resume-cli ats-check [OPTIONS]

Options:
  --job-desc PATH        Path to job description file (required)
  -v, --variant TEXT     Resume variant to check (default: v1.0.0-base)
  --output PATH          Save report as JSON file

Scoring Categories:

• Format Parsing (20pts): Text extraction and formatting checks
• Keywords (30pts): Job keyword matching (uses AI if available, falls back to regex)
• Section Structure (20pts): Standard ATS sections validation
• Contact Info (15pts): Email, phone, location verification
• Readability (15pts): Action verbs, metrics, bullet points

Examples:

# Check ATS score for backend variant
resume-cli ats-check -v v1.1.0-backend --job-desc job-posting.txt

# Save report as JSON
resume-cli ats-check --job-desc job.txt --output ats-report.json

Sample Output:

ATS Score: 81/100 (81%)

Good! Your resume is ATS-friendly with room for improvement.

Category Breakdown:
✓ Format Parsing: 20/20
✗ Keywords: 14/30 - Missing: Kubernetes, CI/CD
✓ Section Structure: 20/20
✓ Contact Info: 15/15
✗ Readability: 12/15

Top Recommendations:
  1. Add these keywords to skills or experience: Kubernetes, CI/CD
  2. Add quantifiable metrics (e.g., 'increased by 30%')

validate

Validate resume.yaml schema and data.

resume-cli validate

generate-package

Generate a complete application package: AI-customized resume + cover letter.

Note: This command automatically uses AI customization when a job description is provided. No --ai flag is needed or available.

resume-cli generate-package [OPTIONS]

Options:
  --job-desc PATH        Path to job description file (required)
  -v, --variant TEXT     Resume variant (default: v1.0.0-base)
  --company TEXT         Company name (overrides extraction from job description)
  --non-interactive      Skip questions, use smart defaults
  --no-cover-letter      Skip cover letter generation
  --output-dir PATH      Output directory (default: config setting)
  --include-github-projects  Auto-select GitHub projects matching job technologies

Output Structure:

output/{company}-{date}/
├── resume.md
├── resume.pdf
├── cover-letter.md
└── cover-letter.pdf

Examples:

# Interactive mode (asks questions for cover letter)
resume-cli generate-package --job-desc job-posting.txt --variant v1.1.0-backend

# Non-interactive mode (uses smart guesses)
resume-cli generate-package --job-desc job.txt --company "Acme Corp" --non-interactive

# With GitHub projects auto-selected based on job technologies (AI is automatic)
resume-cli generate-package --job-desc job-posting.txt --variant v1.2.0-ml_ai --include-github-projects

# Note: No --ai flag needed - AI customization is automatic with --job-desc

apply

Log a job application to tracking CSV.

resume-cli apply COMPANY STATUS [OPTIONS]

Arguments:
  COMPANY     Company name
  STATUS      Application status

Options:
  -r, --role TEXT       Job role/title
  -v, --variant TEXT    Resume variant used
  -s, --source TEXT     Application source
  -u, --url TEXT        Job posting URL
  -n, --notes TEXT      Additional notes

Examples:

# Log new application
resume-cli apply Google applied -r "Senior Backend Engineer"

# Update status
resume-cli apply Google interview -v v1.1.0-backend

# With details
resume-cli apply Stripe offer -r "Staff Engineer" -s "Referral" -u "https://stripe.com/jobs"

Valid statuses:

• applied - Application submitted
• interview - Interview scheduled
• offer - Offer received
• rejected - Application rejected
• withdrawn - Withdrew application

analyze

Show application tracking analytics.

resume-cli analyze

sync-github

Sync GitHub projects to resume.yaml.

resume-cli sync-github [OPTIONS]

Options:
  --months INTEGER  Number of months to look back (default: 3)

Requirements:

• GitHub CLI (gh) installed and authenticated
• Run gh auth login first if needed

linkedin-import

Import LinkedIn profile data into resume.yaml.

resume-cli linkedin-import [OPTIONS]

Options:
  --data-file PATH     Path to LinkedIn exported JSON, CSV, or folder (required)
  --output PATH        Output YAML file path (default: resume.yaml)
  --merge              Merge with existing resume.yaml instead of overwriting
  --dry-run            Preview changes without writing to file

Important: Direct URL import is not supported due to LinkedIn API restrictions. You must export your LinkedIn data first.

To get all your data (experience, education, skills, certifications):

1. Go to https://www.linkedin.com/psettings/member-data
2. Click "Request data archive" (NOT just "Profile" data)
3. Wait for LinkedIn to prepare your download (may take a few minutes to hours)
4. Download the ZIP file containing your data
5. Extract the ZIP file - you can use either:
   • The extracted folder directly (contains Profile.csv, Positions.csv, Education.csv, Skills.csv, etc.)
   • A single JSON file (e.g., profile.json)
   • A single CSV file (e.g., Profile.csv)

Note: The basic "Profile" export (Profile.csv) only contains basic fields (name, headline, summary). It does NOT include experience, education, skills, or certifications. You must request the full data archive to get those.

Examples:

# Import from extracted LinkedIn data folder (recommended for full data)
resume-cli linkedin-import --data-file /path/to/extracted/linkedin-folder/

# Import from exported JSON (creates new resume.yaml)
resume-cli linkedin-import --data-file profile.json

# Merge with existing resume.yaml
resume-cli linkedin-import --data-file profile.json --merge

# Preview changes before importing
resume-cli linkedin-import --data-file profile.json --dry-run

# Import to custom path
resume-cli linkedin-import --data-file profile.json --output my-resume.yaml

Field Mapping:

LinkedIn Field	resume.yaml Field
firstName/lastName	contact.name
email	contact.email
phone	contact.phone
location	contact.location
headline/summary	professional_summary.base
skills	skills.* (auto-categorized)
experience	experience
education	education
certifications	certifications

linkedin-export

Export resume.yaml data to LinkedIn-friendly format.

resume-cli linkedin-export [OPTIONS]

Options:
  -v, --variant TEXT    Resume variant to export (default: v1.0.0-base)
  -o, --output PATH    Output file path (default: output/linkedin-update.txt)
  --format CHOICE      Output format: linkedin, plain (default: linkedin)

Examples:

# Export for LinkedIn update
resume-cli linkedin-export

# Export specific variant
resume-cli linkedin-export -v v1.1.0-backend

# Export to custom file
resume-cli linkedin-export -o my-linkedin-update.txt

# Export plain text format
resume-cli linkedin-export --format plain

Note: LinkedIn has character limits for each field. You may need to trim bullets to fit LinkedIn's format.

init

Initialize resume.yaml from existing resume files.

resume-cli init [OPTIONS]

Options:
  --from-existing    Parse existing resume files

resume.yaml Structure

The single source of truth for all resume data:

meta:
  version: "2.0.0"
  last_updated: "2025-02-03"

contact:
  name: "Your Name"
  credentials: ["P.E."]
  phone: "555-123-4567"
  email: "you@example.com"
  location:
    city: "City"
    state: "ST"
    zip: "12345"
  urls:
    github: "https://github.com/username"
    linkedin: "https://linkedin.com/in/username"

professional_summary:
  base: "Default summary for all variants..."
  variants:
    backend: "Specialized backend summary..."
    ml_ai: "ML/AI focused summary..."

skills:
  programming_languages:
    - { name: "Python", level: "Expert", years: 8 }
  cloud_devops:
    - { name: "AWS", services: ["EKS", "EC2"] }

experience:
  - company: "Company Name"
    title: "Job Title"
    start_date: "2022-04"
    end_date: null  # null = current
    bullets:
      - text: "Achievement description..."
        skills: ["DevOps", "Kubernetes"]
        emphasize_for: ["backend", "devops"]

variants:
  v1.0.0-base:
    description: "General software engineering"
    summary_key: "base"
    skill_sections: ["programming_languages", "cloud_devops"]
    max_bullets_per_job: 4

Variants

v1.0.0-base

General software engineering with balanced focus on all skills.

v1.1.0-backend

Backend & DevOps specialization. Emphasizes backend technologies, cloud infrastructure, databases, and APIs.

v1.2.0-ml_ai

ML/AI specialization. Emphasizes AI/ML frameworks, data engineering, model deployment, and Python/PyTorch.

v1.3.0-fullstack

Full-stack specialization. Emphasizes frontend and backend balance and modern web frameworks.

v1.4.0-devops

DevOps & Infrastructure specialization. Emphasizes Kubernetes, Docker, CI/CD, and cloud platforms.

v1.5.0-leadership

Leadership & Technical Staff. Emphasizes architecture, mentoring, and technical strategy.

AI-Powered Generation

Setup

1. Install AI dependencies:

   pip install -e ".[ai]"

2. Create .env file with API keys:

   cp .env.template .env
   # Edit .env to add your keys

3. Configure API keys in .env:

   For Claude (Anthropic):

   ANTHROPIC_API_KEY=your_key_here
   

   For OpenAI:

   OPENAI_API_KEY=your_key_here
   

   Optional: Custom API endpoints (e.g., for z.ai, OpenRouter, or other proxies):

   ANTHROPIC_BASE_URL=https://api.z.ai/api/anthropic
   OPENAI_BASE_URL=https://your-custom-endpoint.com/v1
   

4. Configure provider in config/default.yaml:

   ai:
     provider: anthropic  # or openai
     model: claude-3-5-sonnet-20241022

Usage

For resume generation with AI:

# Use --ai flag with generate command
resume-cli generate --ai --job-desc job-posting.txt

# AI will:
# 1. Extract key requirements from job description
# 2. Reorder bullets to emphasize relevant experience
# 3. Highlight matching skills
# 4. Keep all content truthful (no fake experience)

For complete application packages (resume + cover letter):

# generate-package automatically uses AI - no --ai flag needed!
resume-cli generate-package --job-desc job-posting.txt --variant v1.1.0-backend

# This generates both AI-customized resume AND cover letter

Troubleshooting AI Errors

"anthropic package not installed"

# Install AI dependencies
pip install -e ".[ai]"

"ANTHROPIC_API_KEY not set"

• Make sure you created a .env file from .env.template
• Verify the API key is set correctly in .env
• python-dotenv loads environment variables from .env automatically

"OPENAI_API_KEY not set"

• Same troubleshooting steps as ANTHROPIC_API_KEY above

API connection errors

• Check your network connection
• Verify API key is valid
• If using custom BASE_URL, verify the endpoint is correct

"AI generation failed, falling back to template"

• This is expected behavior when AI API fails
• The system automatically falls back to fast template generation
• Check API key validity and network connection to use AI features
• Set ai.fallback_to_template: false in config to fail instead of falling back

"No such option: --ai" with generate-package

• The generate-package command does not have an --ai flag
• AI customization is automatic when you provide --job-desc
• Simply run: resume-cli generate-package --job-desc job-posting.txt
• Use resume-cli generate --ai --job-desc ... if you only want an AI-customized resume without cover letter

Configuration

Edit config/default.yaml to customize:

output:
  directory: output
  naming_scheme: "resume-{variant}-{date}.{ext}"
  date_format: "%Y-%m-%d"

generation:
  default_variant: v1.0.0-base
  default_format: md
  max_bullets: 4

ai:
  provider: anthropic
  model: claude-3-5-sonnet-20241022
  fallback_to_template: true
  # Multi-generation with AI Judge
  judge_enabled: true  # Use AI judge to select best of N generations
  num_generations: 3   # Number of versions to generate for judge evaluation
  # Optional: Custom API base URLs (can also be set in .env)
  anthropic_base_url: ""
  openai_base_url: ""

tracking:
  enabled: true
  csv_path: tracking/resume_experiment.csv

cover_letter:
  enabled: true
  formats: [md, pdf]
  smart_guesses: true
  tone: professional
  max_length: 400

github:
  username: your_username
  sync_months: 3
  # Auto-projects feature for generate-package
  max_projects: 3

linkedin:
  # LinkedIn import/export settings
  import_merge_mode: merge      # Options: merge, overwrite
  export_format: linkedin       # Options: linkedin, plain
  output_directory: output     # Default directory for LinkedIn exports

PDF Generation

PDF generation requires LaTeX tools:

Ubuntu/Debian

sudo apt-get install texlive-full

Fedora/RPM-based (Fedora, RHEL, CentOS)

sudo dnf install texlive-scheme-full

macOS

brew install mactex

Or use Pandoc

# Ubuntu/Debian
sudo apt-get install pandoc texlive-xetex

# Fedora/RPM-based
sudo dnf install pandoc texlive-xetex

File Structure

resume-cli/
├── resume.yaml              # Single source resume data (gitignored - personal data)
├── cli/                     # CLI package
│   ├── main.py              # Entry point
│   ├── commands/            # Command implementations
│   ├── generators/          # Template + AI engines
│   ├── integrations/        # Tracking, GitHub, LinkedIn, etc.
│   └── utils/               # YAML parser, schema, config
├── resume_pdf_lib/          # Shared PDF generation library
├── templates/               # Jinja2 templates
│   ├── resume_md.j2
│   ├── resume_tex.j2
│   └── cover_letter_md.j2
├── config/                  # Configuration files
│   ├── default.yaml
│   └── variants.yaml
├── output/                  # Generated resumes (gitignored)
├── tracking/                # Application tracking CSV
├── requirements.txt         # Python dependencies
└── setup.py                 # Package installation

Troubleshooting

"resume.yaml not found"

• Run resume-cli init --from-existing to create from existing files
• Or copy resume.example.yaml to resume.yaml and customize

"anthropic package not installed"

pip install -e ".[ai]"

"ANTHROPIC_API_KEY not set"

• Create .env file from .env.template
• Add your API key to .env
• The system automatically loads variables from .env

"PDF compilation failed"

• Install LaTeX tools (see PDF Generation section above)
• Ensure pdflatex or pandoc is in your PATH

"gh command not found"

• Install GitHub CLI: https://cli.github.com/
• Required for resume-cli sync-github command

Development

Running Tests

pytest

Adding New Variants

1. Add variant to resume.yaml under variants:
2. Optionally add variant-specific summary to professional_summary.variants:
3. Run resume-cli generate -v v1.X.Y-variant

Adding New Template Formats

1. Create Jinja2 template in templates/
2. Add format choice to generate command in cli/main.py
3. Update TemplateGenerator in cli/generators/template.py

License

MIT License - See LICENSE file for details.

Roadmap

Current Version: v2.0.0

Completed Features:

• ✅ Core CLI with template-based generation (Jinja2)
• ✅ Multiple output formats (Markdown, LaTeX, PDF)
• ✅ AI-powered resume tailoring (Claude/OpenAI)
• ✅ Cover letter generation (interactive + non-interactive)
• ✅ Application tracking (CSV-based analytics)
• ✅ GitHub project sync and AI enhancement
• ✅ AI Judge for multi-generation quality selection
• ✅ REST API server for programmatic access
• ✅ Schema validation and configuration management
• ✅ 6 pre-configured resume variants
• ✅ LinkedIn folder-based data archive import

---

Short-Term (1-3 Months)

Quality & Testing

• Comprehensive test coverage (pytest for all modules)
• Integration tests for CLI commands
• Mock testing for AI API calls
• Performance benchmarking for large resume.yaml files

User Experience

• Better error messages and actionable guidance
• Progress indicators for long-running operations
• Resume diff/comparison between variants
• "What changed" log when updating resume.yaml

Template Enhancements

• Additional resume templates (modern, minimalist, academic)
• Cover letter templates with different tones
• User-provided custom template support
• Template preview mode

Documentation

• Interactive tutorials for each command
• Video walkthrough for quick onboarding
• Troubleshooting guide for common issues

---

Medium-Term (3-6 Months)

ATS Optimization

• ATS score checker (parse-friendly format validation)
• Keyword density analysis
• Resume format recommendations for popular ATS systems
• Export to ATS-friendly formats (plain text, docx)

Job Board Integration

• Indeed/LinkedIn job posting parser
• Automated application logging (with manual approval)
• Job search aggregator integration

Advanced Analytics

• Visual dashboard for application metrics
• Response rate heatmaps by company size/location
• Variant performance A/B testing tools
• Export analytics to CSV/JSON for external analysis

Multi-Device Support

• Mobile-responsive web interface
• Desktop application (Electron/Tauri)
• Cross-platform sync via cloud storage
• Offline mode with local AI models

---

Long-Term (6-12+ Months)

Interview Preparation

• AI-generated interview questions based on job description
• Mock interview mode with AI interviewer
• Company research aggregation (news, culture, recent wins)
• STAR method response templates

Networking & Career Growth

• Connection finder (find alumni/connections at target companies)
• Salary research and market data integration
• Offer comparison tool with negotiation suggestions
• Personal branding assistant (LinkedIn, portfolio suggestions)

Enterprise Features

• Team collaboration mode (shared resume reviews)
• Template marketplace for resume designs
• Recruiter dashboard for hiring teams
• API rate limiting and usage analytics

Advanced AI Features

• Multi-language resume generation
• Video resume generation support
• Portfolio project recommendations
• Skill gap analysis vs. job requirements

Platform Expansion

• Browser extension for auto-fill applications
• Mobile app for iOS/Android
• Plugin ecosystem for third-party integrations
• White-label offering for career coaches

---

Requested Features

Help us prioritize! If you'd like to see a feature, please:

1. Check existing GitHub issues
2. Create a new issue with the enhancement label
3. Include use cases and why it matters to you

Most requested (community-driven):

• 📊 ATS optimization tools
• 🌐 Web UI
• 📱 Mobile companion app
• 🤝 Collaboration features

---

Contributing

Contributions welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Submit a pull request

Support

For issues or questions:

• Run resume-cli validate to check for errors
• Check documentation in docs/
• Review tracking/ for application history