Frank Goortani CV MCP Server

A Model Context Protocol (MCP) server specifically designed to serve Frank Goortani's CV information. This server provides structured access to CV data, enabling AI assistants and other MCP-compatible clients to retrieve, search, and present professional information in a standardized format.

Project Overview

Purpose

This project implements an MCP server that exposes Frank Goortani's curriculum vitae information through a standardized interface. The server allows AI assistants and MCP-compatible applications to access structured CV data, including professional profile, skills, work experience, and more.

Architecture Overview

The project is built using the FastMCP framework and follows the Model Context Protocol specification. It's designed with a modular architecture:

- Core Module: Contains data structures, tool implementations, and resource definitions
- Server Module: Provides transport handlers for stdio and HTTP communication
- Deployment: Configured for seamless deployment to Cloudflare Workers

graph TD
    LLM[LLM Systems] -->|MCP Protocol| API[CV MCP API]
    API -->|Tools| CV[CV Data]
    API -->|Resource| MD[CV Markdown]
subgraph "Frank Goortani CV MCP Server"

        API

        subgraph "Data Sources"

            CV

            MD

        end

    end
API -->|Deployment| CFW[Cloudflare Workers]

Key Features

- CV Data Access: Structured access to profile information, skills, interests, and work experience
- Markdown Resource: Full CV available as a markdown resource
- Search Capabilities: Text search across all CV sections
- Company Experience Filtering: Targeted retrieval of experience at specific companies
- Media Access: Direct links to resume PDF and profile picture
- Deployment Ready: Configured for Cloudflare Workers deployment
- Dual Transport: Support for both stdio (local) and HTTP (remote) communication

Resources and Tools

Available Resources

| URI | Description | Type |
|-----|-------------|------|
| cv://frankgoortani | Frank Goortani's complete CV in markdown format | text/markdown |

Available Tools

get_profile

Returns Frank Goortani's profile information including name, title, and professional description.

- Parameters: None
- Returns: JSON object with profile information

Example:

// Request
const result = await client.callTool("get_profile", {});
// Response

{

  "name": "Frank Goortani",

  "title": "Hands on Solution Architect | LLM, Web, Cloud, Mobile, Strategy",

  "certifications": ["TOGAF", "PMP"],

  "email": "frank@goortani.com",

  "description": "Visionary technology executive and AI leader with over 25 years of experience..."

}

get_skills

Returns a list of Frank Goortani's professional skills.

- Parameters: None
- Returns: JSON array of skills

Example:

// Request
const result = await client.callTool("get_skills", {});
// Response

[

  "Distributed Systems, API platforms, Microservices, integrations, Workflow systems",

  "Generative AI, Large Language Models (LLMs), AI agents, AI automation, Machine Learning",

  "Reactive and Functional Programming in Go, Python, Java, Swift, Typescript and JavaScript",

  // ...additional skills

]

get_interests

Returns a list of Frank Goortani's professional interests.

- Parameters: None
- Returns: JSON array of interests

Example:

// Request
const result = await client.callTool("get_interests", {});
// Response

[

  "Startups", "GoLang", "Python", "Typescript", "LangChain", "LLMs", "Microservices", "MCPs"

]

search_cv

Searches throughout the CV for specific terms and returns matching sections.

- Parameters:
- query (string): The search term to look for in the CV
- Returns: JSON object with matching sections

Example:

// Request
const result = await client.callTool("search_cv", {
  query: "generative ai"
});
// Response

{

  "matches": [

    {

      "section": "profile",

      "content": "Visionary technology executive and AI leader with over 25 years of experience driving strategic innovation in generative AI..."

    },

    {

      "section": "skills",

      "content": "Generative AI, Large Language Models (LLMs), AI agents, AI automation, Machine Learning"

    }

    // ...additional matches

  ]

}

get_company_experience

Retrieves work experience at a specific company.

- Parameters:
- company (string): Company name to get experience for
- Returns: JSON object with matching experiences. Throws UserError if none found.

Example:

// Request
const result = await client.callTool("get_company_experience", {
  company: "Uber"
});
// Response

{

  "experiences": [

    {

      "company": "Uber",

      "period": "2021-now",

      "title": "Solution Architect",

      "highlights": [

        "Worked on UDE (User Data Extraction) and DSAR (Data Subject Access Request) Automation...",

        // ...additional highlights

      ]

    }

  ]

}

get_resume_link

Returns the URL to Frank Goortani's resume PDF.

- Parameters: None
- Returns: JSON object with resume URL

Example:

// Request
const result = await client.callTool("get_resume_link", {});
// Response

{

  "resumeLink": "media/Frank Goortani Resume--solution-architect-2024.pdf"

}

get_profile_picture

Returns the URL to Frank Goortani's profile picture.

- Parameters: None
- Returns: JSON object with profile picture URL

Example:

// Request
const result = await client.callTool("get_profile_picture", {});
// Response

{

  "pictureLink": "media/frankgoortani.png"

}

Deployment Instructions

Prerequisites

- Bun or Node.js (v18+)
- Wrangler CLI (for Cloudflare deployment)
- Cloudflare account (for deployment)

Local Development

1. Clone the repository

   git clone <repository-url>

   cd cv-mcp

   

2. Install dependencies

   # Using Bun (recommended)

   bun install

   # OR using npm

   npm install

   

3. Start the server locally with stdio transport (for CLI tools)

   # Using Bun

   bun start

   # OR using npm

   npm start

   

4. Start the server locally with HTTP transport (for web applications)

   # Using Bun

   bun run start:http

   # OR using npm

   npm run start:http

   

5. For development with auto-reload

   # Using Bun with stdio transport

   bun run dev

   # Using Bun with HTTP transport

   bun run dev:http

   

Cloudflare Deployment

1. Login to your Cloudflare account

   wrangler login

   

2. Update wrangler.toml if needed

The configuration is already set up in the wrangler.toml file. You may want to customize:
- Custom domain routes
- Environment variables
- KV namespace connections

3. Deploy to Cloudflare

   # Deploy to production

   npm run deploy

# OR deploy to development environment

   npm run deploy:dev

   

4. For local testing of Cloudflare Worker

   npm run dev:cf

   

Environment Configuration

The project uses the following environment variables that can be configured:

- PORT: HTTP port for local development (default: 3001)
- HOST: Host binding for HTTP server (default: 0.0.0.0)
- ENVIRONMENT: Current environment (development/production)
- API_KEY: API key required in the x-api-key header for all requests
- REQUEST_TIMEOUT: Maximum request duration in milliseconds (default: 120000)
- PING_INTERVAL: Interval between ping messages in milliseconds (default: 60000)

In Cloudflare, these variables are configured in the wrangler.toml file:

[vars]
ENVIRONMENT = "production"
[env.dev.vars]

ENVIRONMENT = "development"

Usage Examples

Connecting to the Server

Connecting from a CLI MCP Client

```bash