Winning Cv

An MCP (Model Context Protocol) server for AI-powered job matching and CV generation. Search jobs on LinkedIn and Seek, match them against your CV, get detailed fit analysis, and generate tailored CVs.

Features

• 🔍 search_jobs - Search for jobs on LinkedIn and/or Seek
• 📊 match_jobs - AI-powered CV matching with scoring (1-10)
• 🎯 analyze_job - Deep-dive job analysis with fit assessment, strengths, gaps, red flags, and cover letter angles
• 📝 generate_cv - Generate tailored CVs optimized for specific jobs

Installation

npm install
npm run build

Configuration

Copy .env.example to .env and configure:

cp .env.example .env

Environment Variables

Variable	Description	Default
OPENAI_API_KEY	OpenAI API key (or "dummy" for proxies)	Required
OPENAI_BASE_URL	OpenAI-compatible API endpoint	OpenAI default
OPENAI_MODEL	Model to use for AI features	gpt-4o-mini

Using with GitHub Copilot Proxy

export OPENAI_API_KEY=dummy
export OPENAI_BASE_URL=http://localhost:14141/v1
export OPENAI_MODEL=gpt-4o

Usage

Running the Server

node dist/index.js

MCP Integration

Add to your MCP client configuration:

{
  "mcpServers": {
    "winning-cv": {
      "command": "node",
      "args": ["/path/to/winning-cv-mcp/dist/index.js"],
      "env": {
        "OPENAI_API_KEY": "your-key",
        "OPENAI_MODEL": "gpt-4o"
      }
    }
  }
}

Tools

search_jobs

Search for jobs on LinkedIn and/or Seek.

Input:

{
  "keywords": "Digital Transformation Manager",
  "location": "Melbourne, Australia",
  "sources": ["linkedin", "seek"],
  "days_old": 7
}

Output: List of jobs with title, company, location, URL, salary, and posted date.

match_jobs

Score how well your CV matches job postings.

Input:

{
  "cv_path": "/path/to/your-cv.pdf",
  "job_urls": [
    "https://www.seek.com.au/job/123456",
    "https://www.linkedin.com/jobs/view/789012"
  ],
  "threshold": 7
}

Output: Jobs meeting the threshold with scores, strengths, and gaps.

analyze_job

Get detailed analysis of a job against your CV.

Input:

{
  "cv_path": "/path/to/your-cv.pdf",
  "job_url": "https://www.seek.com.au/job/123456"
}

Output:

• Fit assessment (score 1-10, recommendation: strong_apply/apply/consider/skip)
• Strengths with categories
• Gaps with severity (critical/moderate/minor) and mitigation suggestions
• Red flags
• Cover letter angles with talking points
• Key requirements checklist (met/unmet with evidence)

generate_cv

Generate a tailored CV for a specific job.

Input:

{
  "cv_path": "/path/to/your-cv.pdf",
  "job_url": "https://www.seek.com.au/job/123456",
  "output_format": "markdown",
  "tone": "professional"
}

Options:

• output_format: markdown, plain_text, or structured (JSON)
• tone: professional, creative, technical, or executive
• focus_areas: Optional array of areas to emphasize

Output: Tailored CV content with tailoring notes (keywords emphasized, skills highlighted, experience reframed).

Supported CV Formats

• PDF
• DOCX
• TXT
• Markdown (.md)

Dependencies

• Node.js 18+
• Playwright (for web scraping)

Installing Playwright Browsers

npx playwright install chromium

Development

# Build
npm run build

# Watch mode
npm run dev

# Test a tool
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | node dist/index.js

License

MIT

Author

Jack Hui (https://jackhui.com.au)