Developer Portfolio MCP Server
An MCP server that lets recruiters and AI assistants explore your profile, skills, and projects through natural language.
What is this?
This project is a Model Context Protocol (MCP) server — a standardized bridge between AI assistants (Cursor, Claude Desktop, etc.) and your developer portfolio data.
Instead of sending recruiters a static PDF resume, you give them (or their AI tools) a live MCP server they can query interactively:
- "What Python projects has this developer built?"
- "How well does their stack match this job description?"
- "Generate an interview prep brief for a Senior Engineer role."
Why recruiters care: It demonstrates you understand modern AI tooling, can build production integrations, and think about developer experience.
Quick Start
1. Clone and install
git clone https://github.com/YOUR_USERNAME/developer-portfolio-mcp.git
cd developer-portfolio-mcp
uv sync --extra dev
2. Customize your data
Edit these JSON files with your real information:
| File | What to update |
|---|---|
src/developer_portfolio_mcp/data/profile.json |
Name, email, GitHub, LinkedIn, summary |
src/developer_portfolio_mcp/data/projects.json |
Your real projects |
src/developer_portfolio_mcp/data/skills.json |
Your actual skills |
3. Run the server
uv run developer-portfolio-mcp
4. Connect to Cursor
Copy the example config and update the path:
cp .cursor/mcp.json.example .cursor/mcp.json
# Edit the absolute path in mcp.json to your project directory
Then in Cursor: Settings → MCP — your server should appear. Try asking:
"Use the developer-portfolio MCP to show me this candidate's top projects."
What's Inside
Tools (6) — Actions the AI can perform
| Tool | Description |
|---|---|
get_developer_profile |
Core profile and availability |
list_projects |
Portfolio projects (filterable by status) |
get_project_details |
Deep dive on one project |
search_skills |
Search skills by keyword/category |
get_contact_info |
Contact links and scheduling info |
compare_stack |
Match job requirements vs. your skills |
Resources (3) — Read-only data endpoints
| URI | Content |
|---|---|
profile://resume |
Full resume in Markdown |
profile://skills/{category} |
Skills by category |
projects://list |
All projects as JSON |
Prompts (3) — Reusable AI templates
| Prompt | Use case |
|---|---|
interview_prep |
Generate role-specific interview prep |
recruiter_brief |
One-page candidate summary |
project_pitch |
Pitch a project to a hiring manager |
Project Structure
developer-portfolio-mcp/
├── src/developer_portfolio_mcp/
│ ├── server.py # MCP server (tools, resources, prompts)
│ ├── loaders.py # Data loading utilities
│ └── data/ # Your portfolio JSON (customize these!)
│ ├── profile.json
│ ├── projects.json
│ └── skills.json
├── tests/ # Unit tests
├── docs/
│ └── LEARNING_GUIDE.md # MCP basics → advanced (start here to learn!)
├── .cursor/mcp.json.example
├── pyproject.toml
└── README.md
Run Tests
uv run pytest -v
Learning MCP
New to MCP? Read docs/LEARNING_GUIDE.md — a structured path from basics to advanced concepts, using this project as your hands-on lab.
Topics covered:
- What MCP is and why it matters
- Tools vs Resources vs Prompts
- Transports (stdio, HTTP)
- Building your own tools
- Deploying for recruiters
Push to GitHub
# After customizing your data:
git add .
git commit -m "Add developer portfolio MCP server"
git remote add origin https://github.com/YOUR_USERNAME/developer-portfolio-mcp.git
git push -u origin main
Add this to your resume/LinkedIn:
Built an MCP server exposing developer portfolio data to AI assistants — github.com/YOUR_USERNAME/developer-portfolio-mcp
Tech Stack
- MCP Python SDK (FastMCP)
- Python 3.11+
- uv for package management
License
MIT — use freely, customize for your portfolio, and share with recruiters.
