cvmaker-mcp
A local MCP server that bridges AI assistants with the CV Maker app to generate pixel-perfect CV PDFs.
README
cvmaker-mcp
A local Model Context Protocol (MCP) server that bridges AI assistants (Claude Desktop, Cursor, etc.) with the CV Maker app to generate pixel-perfect CV PDFs.
How it works
- You describe your background to an AI assistant (e.g. Claude Desktop)
- The AI structures your information and calls this MCP tool
- The AI will explicitly ask where you want the final PDF saved, ensuring total control over the output destination.
- A local headless browser securely visits resume.nigamelastic.com, computationally injects your CV data locally (without transmitting to any backend server), and captures the resume in PDF format.
- In addition to PDFs, the AI can securely save or load your structured "profiles" as
.cv.jsonfiles directly to your hard drive, bypassing cloud storage entirely.
🔒 Privacy first: Your CV data is injected into a local browser's localStorage and is never sent to any external server.
Prerequisites
- Node.js v18 or later
- An MCP-compatible AI client:
- Claude Desktop
- Cursor
- Any client supporting the MCP stdio transport
Quick Start (No Install)
You can run the MCP server directly using npx in your MCP client configuration (Claude Desktop, Cursor, etc.):
{
"mcpServers": {
"cvmaker": {
"command": "npx",
"args": ["-y", "@nigamelastic/cvmaker-mcp"]
}
}
}
[!TIP] Smart Browser Detection: The first time you run this, it will check for a system browser (Chrome/Brave/Chromium). If none is found, it will automatically download a lightweight version (~130MB) and notify you via logs.
Prerequisites
- Node.js v18 or later
- An MCP-compatible AI client:
- Claude Desktop
- Cursor
- Any client supporting the MCP stdio transport
Configuration
Claude Desktop
Add the following to your Claude Desktop config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
{
"mcpServers": {
"cvmaker": {
"command": "npx",
"args": ["-y", "@nigamelastic/cvmaker-mcp"]
}
}
}
Development / Manual Installation
If you prefer to run from source:
git clone https://github.com/nigamelastic/cvmaker_mcp.git
cd cvmaker_mcp
npm install
node src/index.js
Usage
Once configured and Claude Desktop is restarted, you can use natural language:
"Create a CV for a Senior React Developer with 8 years of experience at Google and Meta. Use the tech template."
"Here's my LinkedIn summary: [paste text]. Build me a CV PDF using the elegant template."
Available Templates
| Template | Description |
|---|---|
standard |
Clean, traditional single-column layout |
modern |
Contemporary design with accent colours |
minimal |
Ultra-clean, whitespace-focused |
elegant |
Sophisticated serif-accented layout |
sidebar |
Two-column with a dark sidebar |
tech |
Developer-focused, monospace accents |
europass |
Traditional, highly structured European standardized layout |
executive |
Formal, dense layout optimized for extensive senior experience |
Tool Reference
generate_cv_pdf
Generates a pixel-perfect A4 PDF using the CV Maker engine natively.
| Parameter | Type | Required | Description |
|---|---|---|---|
cv.personal |
object | ✅ | Name, title, email, phone, website, location, summary |
cv.experience |
array | ✅ | List of { role, company, period, description } |
cv.education |
array | ✅ | List of { degree, school, period } |
cv.skills |
array | ✅ | List of skill strings |
cv.customSections |
array | — | Up to 4 custom { title, content } sections |
cv.activeTemplate |
string | — | Template name (default: standard) |
destination_dir |
string | ✅ | Absolute directory path where you want the PDF saved |
output_filename |
string | — | PDF filename without extension |
Returns: File path to the generated PDF on your local machine.
export_cv_json
Saves the structured CV JSON to a local file in ~/.cvmaker-profiles/. Perfect for saving "profiles" to iterate on later.
| Parameter | Type | Required | Description |
|---|---|---|---|
cv |
object | ✅ | structured CV JSON Payload |
output_filename |
string | — | Desired profile filename without extension |
import_cv_json
Loads a previously saved CV JSON profile from your hard drive structure back into the AI context.
| Parameter | Type | Required | Description |
|---|---|---|---|
file_path |
string | ✅ | Absolute absolute path to the .cv.json file |
extract_resume_data
A helper tool for validating AI-extracted resume/LinkedIn text explicitly against the strict CV JSON Schema before doing heavier rendering operations.
| Parameter | Type | Required | Description |
|---|---|---|---|
cv |
object | ✅ | structured CV JSON Payload |
get_available_templates
Fetches a list of valid template string IDs supported by CV Maker without requiring manual code inspection.
Warranty and Liability Disclaimer
This software is provided "as is", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose and noninfringement. In no event shall the authors or copyright holders be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the software or the use or other dealings in the software.
Use of third-party site: This MCP server interacts with https://resume.nigamelastic.com/. While this server sanitizes data locally before injection, the user acknowledges that they are responsible for the content they process and that the final PDF rendering occurs within a headless browser instance.
License
MIT © nigamelastic
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。