Skip to main content
Glama
mcp-cv

mcp-cv

by 123justin123

mcp-cv

npm

My resume, queryable from your AI assistant.

A Model Context Protocol server that exposes my background, projects and contact details. Plug it into Claude, Cursor or any MCP client and just ask: "Who is Justin?", "Has he used Kafka?", "How can I reach him?".

npx -y @just1-dev/mcp-cv

Installation

Claude Desktop

In claude_desktop_config.json:

{
  "mcpServers": {
    "justin": {
      "command": "npx",
      "args": ["-y", "@just1-dev/mcp-cv"]
    }
  }
}

Claude Code

claude mcp add justin -- npx -y @just1-dev/mcp-cv

Cursor, VS Code and other clients

Same idea: command npx, arguments -y @just1-dev/mcp-cv, stdio transport.

Related MCP server: Bijon Portfolio MCP Server

What the server exposes

Tools

Tool

Input

Purpose

get_profile

none

Summary, tech stack, availability, work experience, education

get_projects

tech?, slug?

Project case studies, optionally narrowed to one technology or one project

get_contact

none

Email, LinkedIn, GitHub, website

All three are read-only, declare an outputSchema and return validated structuredContent.

Resources

  • justin://resume: the whole resume

  • justin://projects/{slug}: a single project, with listing and slug completion

Architecture

src/
├── index.ts          CLI entry point, serveStdio
├── server.ts         createServer(): wires tools and resources
├── schemas.ts        Zod schemas: types, validation and output schemas
├── data/cv.ts        single source of truth for the data
├── tools/index.ts    the three tools
├── resources.ts      resources and URI template
└── lib/
    ├── cv.ts         public view of the resume, age computation
    └── result.ts     MCP result helper

Two MCP primitives serve the same data through two channels: tools are called by the model, resources are attached by the user.

The server holds no model and no search: it serves structured data and instructs the client's model to answer only from it. createServer() is transport-agnostic: index.ts serves it over stdio, and the test suite drives it through a real MCP client and createMcpHandler, in process, on the protocol revision hosts negotiate today.

Development

npm install
npm run dev        # run the server from source
npm test           # node:test suite
npm run typecheck
npm run inspect    # build, then open the MCP Inspector in the browser

Available Tools

3 tools
get_contactContactA
Read-onlyIdempotent

How to reach Justin: email, LinkedIn, GitHub, website.

Inputs: no named parameters. Returns: email, github, website, linkedin

get_profileProfileA
Read-onlyIdempotent

Justin's background: summary, tech stack, availability, work experience and education.

Inputs: no named parameters. Returns: profile, education, experience

get_projectsProjectsA
Read-onlyIdempotent

Justin's projects as case studies, filterable by technology or by slug.

Inputs: slug, tech. Returns: projects

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.1/5.0

    Scored across 3 tools

    Disambiguation5/5

    Each tool targets a clearly distinct section of a CV: professional background, projects, and contact information. There is no overlap or ambiguity between them.

    Naming Consistency5/5

    All tool names follow a uniform get_<noun> pattern, making the API predictable and easy to navigate. The naming is fully consistent and intuitive.

    Tool Count5/5

    Three tools is a well-scoped size for a personal CV server, with each tool covering a natural top-level section. The count is neither too thin nor excessive.

    Completeness5/5

    The tool surface covers the essential aspects of a CV: profile (including experience and education), projects, and contact. There are no obvious missing operations that an agent would need to get a complete picture of the candidate.

    Maintenance

    ActivitySlowing
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      Enables searching and retrieving portfolio data including experience, skills, and contact information through natural language queries.
      5
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Exposes a structured professional resume as a set of AI-queryable tools, enabling AI clients like Claude Desktop to query summary, experience, skills, projects, and tailor resumes to job descriptions.
      1
      MIT