CV Resume Builder MCP
Automatically generates and updates CVs/resumes by aggregating data from git commits, Jira tickets, Credly certifications, and existing PDF resumes, with LaTeX formatting support.
README
CV Resume Builder MCP
AI-powered CV and resume builder using Model Context Protocol (MCP)
Automatically generate and update your CV/resume from git commits, Jira tickets, Credly certifications, and LinkedIn. Built for professionals who want their CV to stay current without manual updates. Generates ATS-compliant LaTeX CVs that pass Applicant Tracking Systems.
Features
- 📊 Git commits - Track your code contributions automatically ✅
- 🎫 Jira tickets - Pull completed projects and tasks ⚠️ (requires testing)
- 🏆 Credly badges - Sync certifications and achievements ✅
- 💼 LinkedIn profile - Not implemented yet 🚧 (authentication required)
- 📄 PDF parsing - Extract content from existing CVs ✅
- 🚀 Enhanced CV generation - Combine all data sources ✅
- 📝 LaTeX support - Generate professional, ATS-compliant CVs ✅
- 🤖 ATS-friendly - Clean formatting that passes Applicant Tracking Systems ✅
Quick Start
Prerequisites
- Python 3.10+
uvinstalled (for uvx):curl -LsSf https://astral.sh/uv/install.sh | sh- An MCP-compatible AI assistant (Claude Desktop, Kiro, etc.)
Installation
Using uvx (recommended - no installation needed!):
Just configure your MCP client and uvx handles the rest.
Or install with pip:
pip install cv-resume-builder-mcp
Note: The configurations below use
uvxwhich automatically downloads and runs the latest version from PyPI. If you're developing locally or want to test unreleased features, see TESTING.md for local development setup.
Configuration
For Claude Desktop
Edit ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"cv-resume-builder": {
"command": "uvx",
"args": ["cv-resume-builder-mcp"],
"env": {
"AUTHOR_NAME": "your-git-username",
"REPOS": "default:/absolute/path/to/your-repo"
}
}
}
}
For multiple repositories: Change REPOS to: "CompanyA:/path1,CompanyB:/path2,Personal:/path3"
For Kiro IDE
Edit ~/.kiro/settings/mcp.json:
{
"mcpServers": {
"cv-resume-builder": {
"command": "uvx",
"args": ["cv-resume-builder-mcp"],
"env": {
"AUTHOR_NAME": "your-git-username",
"REPOS": "default:/absolute/path/to/your-repo"
}
}
}
}
For multiple repositories: Change REPOS to: "CompanyA:/path1,CompanyB:/path2,Personal:/path3"
Important: Use absolute paths (no ~). Get it with pwd in your repo directory.
Restart Your AI Assistant
After configuration, restart Claude Desktop or Kiro to load the MCP server.
Test It
"List available MCP tools"
You should see tools like get_git_log, read_cv, parse_cv_pdf, etc.
Usage Examples
"Get my git commits from the last 6 months and suggest CV updates"
With multiple repositories:
"List all my configured repositories"
"Get my commits from CompanyA for the last 3 months"
"Show me all my work across all repositories in the last year"
Parse existing CV
"Parse my CV at ~/Documents/resume.pdf"
Generate enhanced CV
"Generate an enhanced CV using my existing resume.pdf and recent work from the last 3 months"
Get certifications
"Get my Credly badges and add them to my CV"
Analyze commit impact (NEW!)
"Analyze my commits from the last month and show me what I actually built"
"Get detailed code changes for commit abc123 to understand the impact"
"Show me the stats for my recent commits to highlight achievements"
Career guidance
"Based on my commits, Jira tickets, and certifications, what seniority level am I at?"
"Analyze my work and suggest what skills I need for senior/staff/principal level"
"Review my achievements and help me position myself for a promotion"
"What's missing from my profile to reach the next level?"
Optional Integrations
Add these to your MCP configuration's env section:
Jira (⚠️ Requires Testing)
"JIRA_URL": "https://your-company.atlassian.net",
"JIRA_EMAIL": "your-email@example.com",
"JIRA_API_TOKEN": "your-api-token",
"JIRA_USER": "your-email@example.com"
Get API token: https://id.atlassian.com/manage-profile/security/api-tokens
Note: Jira integration is functional but requires more testing across different Jira configurations. Please report any issues!
Credly (✅ Fully Tested)
"CREDLY_USER_ID": "your-credly-username"
Find your username in your Credly profile URL: https://www.credly.com/users/YOUR-USERNAME
LinkedIn (🚧 Not Implemented)
"LINKEDIN_PROFILE_URL": "https://www.linkedin.com/in/yourprofile"
Note: LinkedIn integration is not yet implemented due to authentication requirements. The tool currently only returns your profile URL. For now, manually copy your LinkedIn achievements to wins.md file. Contributions welcome!
CV Formatting
"MAX_BULLETS_PER_EXPERIENCE": "5"
Available Tools
| Tool | Description |
|---|---|
get_git_log |
Get your git commits from default repo (excludes merge commits) |
list_repos |
List all configured repositories |
get_git_log_by_repo |
Get commits from a specific repository |
get_git_log_all_repos |
Get commits from all repos, grouped by project |
get_commit_details |
NEW! Get detailed commit info including code changes (diff) for impact analysis |
analyze_commits_impact |
NEW! Analyze multiple commits with stats to understand actual work done |
read_cv |
Read your current LaTeX CV |
read_wins |
Read your wins.md achievements file |
get_jira_tickets |
Get completed Jira tickets |
get_credly_badges |
Get your certifications from Credly |
get_linkedin_profile |
Read your LinkedIn profile summary |
parse_cv_pdf |
Extract text from existing CV/resume PDF |
generate_enhanced_cv |
Combine all data sources into comprehensive report |
get_cv_guidelines |
Get formatting rules and constraints |
Project Structure
cv-resume-builder-mcp/
├── src/cv_resume_builder_mcp/
│ ├── __init__.py
│ └── server.py # Main MCP server
├── tests/
│ └── test_server.py # Unit tests
├── pyproject.toml # Package configuration
├── requirements.txt # Dependencies
├── .env.example # Configuration template
├── cv.tex # LaTeX CV template
├── wins.md # Manual achievements tracking
└── README.md # This file
Development
For End Users (Production)
Use the uvx configuration shown above. It automatically downloads the latest stable version from PyPI.
For Contributors (Local Development)
Just testing locally? Clone the repo:
git clone https://github.com/eyaab/cv-resume-builder-mcp.git
cd cv-resume-builder-mcp
pip install -e ".[dev]"
Want to contribute changes? Fork first, then clone your fork:
- Click "Fork" on GitHub
- Clone your fork:
git clone https://github.com/YOUR-USERNAME/cv-resume-builder-mcp.git
cd cv-resume-builder-mcp
pip install -e ".[dev]"
Then use the local development configuration from TESTING.md which uses python3 -m instead of uvx.
Run Tests
pytest
Format Code
black src/
ruff check src/
Contributing
Contributions welcome! Please:
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Ensure no credentials are hardcoded
- Submit a pull request
See CONTRIBUTING.md for details.
Troubleshooting
MCP server not showing up
- Verify absolute paths in config (use
pwd) - Restart your AI assistant completely
- Check for typos in configuration
Git log returns empty
- Check your git author name:
git config user.name - Update
AUTHOR_NAMEto match exactly
"Command not found: uvx"
- Install uv:
curl -LsSf https://astral.sh/uv/install.sh | sh - Or:
brew install uv
Jira/Credly errors
- Verify API tokens are correct
- Check URLs don't have trailing slashes
- Ensure services are accessible
Security
- All credentials stored in environment variables
- No data sent to external services except configured integrations
- Git history stays local
- Open source - audit the code yourself
License
MIT License - see LICENSE file for details.
Support
- 🐛 Report bugs - Open an issue on GitHub
- 💡 Request features - Create a feature request issue
- 💬 Ask questions - Start a discussion on GitHub
Keywords: cv builder, resume builder, mcp, model context protocol, ai resume, ai cv, automatic cv, career tracker, latex cv, resume generator, cv generator, developer resume, tech resume, ats compliant, ats friendly, applicant tracking system
Made with ❤️ for developers who hate updating their CVs manually
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。