portfolio-mcp-server
Exposes 32 portfolio tools over stdio so MCP-compatible AI clients can read and manage profile, projects, skills, certifications, education, experience, and testimonials from a universal-portfolio instance.
README
portfolio-mcp-server
An MCP (Model Context Protocol) server that connects AI tools — Claude Desktop, Cursor, GitHub Copilot — directly to your universal-portfolio instance.
What it does
Exposes all 32 portfolio tools over stdio so any MCP-compatible AI client can:
- Read your profile, projects, skills, certifications, education, experience, and testimonials
- Create, update, and delete any portfolio item (with a write-enabled API key)
Related
- universal-portfolio — the full-stack portfolio app this server connects to: github.com/muhammad-awais-web-dev/universal-portfolio
Setup
1. Install dependencies
npm install
2. Configure environment
cp .env.example .env
Edit .env:
# URL of your running portfolio (local or deployed)
PORTFOLIO_URL=https://your-portfolio-domain.com
# API key from portfolio admin → Settings → API Keys
# Enable "can_write" on the key for write tool access
MCP_API_KEY=your_api_key_here
3. Test manually
npm start
For development with auto-reload:
npm run dev
The server communicates over stdio — it won't print anything until an MCP client connects. Startup messages appear on stderr.
Connect to Claude Desktop
Edit ~/.config/claude/claude_desktop_config.json:
{
"mcpServers": {
"portfolio": {
"command": "npx",
"args": ["tsx", "/path/to/portfolio-mcp-server/src/index.ts"],
"env": {
"PORTFOLIO_URL": "https://your-portfolio-domain.com",
"MCP_API_KEY": "your_api_key_here"
}
}
}
}
Restart Claude Desktop. You'll see portfolio listed under MCP tools.
Connect to Cursor
Edit ~/.cursor/mcp.json:
{
"mcpServers": {
"portfolio": {
"command": "npx",
"args": ["tsx", "/path/to/portfolio-mcp-server/src/index.ts"],
"env": {
"PORTFOLIO_URL": "https://your-portfolio-domain.com",
"MCP_API_KEY": "your_api_key_here"
}
}
}
}
Connect to VS Code
Requires VS Code 1.99+ with the GitHub Copilot extension.
Create or edit .vscode/mcp.json in your workspace (or add to your user settings.json under "mcp"):
{
"servers": {
"portfolio": {
"type": "stdio",
"command": "npx",
"args": ["tsx", "/path/to/portfolio-mcp-server/src/index.ts"],
"env": {
"PORTFOLIO_URL": "https://your-portfolio-domain.com",
"MCP_API_KEY": "your_api_key_here"
}
}
}
}
Open the Copilot Chat panel, switch to Agent mode, and the portfolio server will appear under available tools.
Connect to GitHub Copilot CLI
Edit ~/.copilot/mcp-config.json:
{
"mcpServers": {
"portfolio": {
"type": "stdio",
"command": "npx",
"args": ["tsx", "/path/to/portfolio-mcp-server/src/index.ts"],
"env": {
"PORTFOLIO_URL": "https://your-portfolio-domain.com",
"MCP_API_KEY": "your_api_key_here"
}
}
}
}
Available Tools (32 total)
Read tools — any API key
| Tool | Description |
|---|---|
get_profile |
Profile info, bio, social links |
list_projects |
Published projects (filter by category/skill, paginated) |
get_project |
Single project by ID or slug |
list_skills |
All skills (filter by category, paginated) |
get_skill |
Single skill by ID or name |
list_certifications |
All certifications |
get_certification |
Single certification by ID |
list_education |
Education history |
get_education |
Single education entry by ID |
list_experience |
Work experience |
get_experience |
Single experience entry by ID |
list_testimonials |
Testimonials (filter featured) |
get_testimonial |
Single testimonial by ID |
Write tools — API key with can_write = true
| Tool | Description |
|---|---|
update_profile |
Update name, bio, email, social links, etc. |
create_project |
Create a new project |
update_project |
Update project by ID (including body_html) |
delete_project |
Delete project by ID |
create_skill |
Create a new skill |
update_skill |
Update skill by ID |
delete_skill |
Delete skill by ID |
create_certification |
Create a new certification |
update_certification |
Update certification by ID |
delete_certification |
Delete certification by ID |
create_education |
Create an education entry |
update_education |
Update education by ID |
delete_education |
Delete education by ID |
create_experience |
Create a work experience entry |
update_experience |
Update experience by ID |
delete_experience |
Delete experience by ID |
create_testimonial |
Create a testimonial |
update_testimonial |
Update testimonial by ID |
delete_testimonial |
Delete testimonial by ID |
Example prompts (once connected to Claude/Cursor/Copilot)
- "List all my projects"
- "What skills do I have in the Backend category?"
- "Update my profile bio to: I'm a full-stack developer..."
- "Create a new skill called Docker in the DevOps category"
- "Update project 5 — set body_html to:
<p>New description</p>" - "Publish project with slug 'my-app'"
Requirements
- Node.js 18+
- A running universal-portfolio instance
- An MCP API key (generated in the portfolio admin panel)
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。