MCP Builder

This skill helps you create Model Context Protocol (MCP) servers to extend GitHub Copilot and AI agent capabilities with custom tools, resources, and data sources.

When to Use This Skill

Use this skill when you need to:
- Extend AI agents with custom functionality
- Connect AI to proprietary data sources
- Build reusable tools for AI workflows
- Create integrations with external APIs
- Implement custom prompts and resources for AI agents
- Enable AI agents to interact with domain-specific systems

What is MCP?

Model Context Protocol (MCP) is an open protocol that standardizes how AI applications like GitHub Copilot connect to data sources and tools. It allows you to:
- Expose tools: Functions that GitHub Copilot and AI agents can call
- Provide resources: Data that AI can access
- Define prompts: Reusable prompt templates
- Stream data: Real-time updates to AI agents

Quick Start

Installation

# Create new MCP server
npm init -y
npm install @modelcontextprotocol/sdk

# Or with TypeScript
npm install --save-dev typescript @types/node
npx tsc --init

Basic MCP Server

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
  CallToolRequestSchema,
  ListToolsRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

// Create server
const server = new Server(
  {
    name: "example-server",
    version: "1.0.0",
  },
  {
    capabilities: {
      tools: {},
    },
  }
);

// Register tool handlers
server.setRequestHandler(ListToolsRequestSchema, async () => {
  return {
    tools: [
      {
        name: "get-weather",
        description: "Get current weather for a location",
        inputSchema: {
          type: "object",
          properties: {
            location: {
              type: "string",
              description: "City name or zip code",
            },
          },
          required: ["location"],
        },
      },
    ],
  };
});

server.setRequestHandler(CallToolRequestSchema, async (request) => {
  if (request.params.name === "get-weather") {
    const { location } = request.params.arguments as { location: string };

    // Mock weather data (replace with actual API call)
    const weather = {
      location,
      temperature: 72,
      condition: "Sunny",
      humidity: 45,
    };

    return {
      content: [
        {
          type: "text",
          text: JSON.stringify(weather, null, 2),
        },
      ],
    };
  }

  throw new Error(`Unknown tool: ${request.params.name}`);
});

// Start server
async function main() {
  const transport = new StdioServerTransport();
  await server.connect(transport);
  console.error("Weather MCP server running on stdio");
}

main().catch((error) => {
  console.error("Server error:", error);
  process.exit(1);
});

Building Tools

Tool with External API

import axios from "axios";

// GitHub repo info tool
server.setRequestHandler(ListToolsRequestSchema, async () => {
  return {
    tools: [
      {
        name: "get-repo-info",
        description: "Get information about a GitHub repository",
        inputSchema: {
          type: "object",
          properties: {
            owner: {
              type: "string",
              description: "Repository owner (username or organization)",
            },
            repo: {
              type: "string",
              description: "Repository name",
            },
          },
          required: ["owner", "repo"],
        },
      },
    ],
  };
});

server.setRequestHandler(CallToolRequestSchema, async (request) => {
  if (request.params.name === "get-repo-info") {
    const { owner, repo } = request.params.arguments as {
      owner: string;
      repo: string;
    };

    try {
      const response = await axios.get(
        `https://api.github.com/repos/${owner}/${repo}`
      );

      const info = {
        name: response.data.name,
        description: response.data.description,
        stars: response.data.stargazers_count,
        forks: response.data.forks_count,
        language: response.data.language,
        url: response.data.html_url,
      };

      return {
        content: [
          {
            type: "text",
            text: JSON.stringify(info, null, 2),
          },
        ],
      };
    } catch (error) {
      return {
        content: [
          {
            type: "text",
            text: `Error fetching repository: ${error.message}`,
          },
        ],
        isError: true,
      };
    }
  }

  throw new Error(`Unknown tool: ${request.params.name}`);
});

Tool with Database Access

import { Pool } from "pg";

const pool = new Pool({
  host: process.env.DB_HOST,
  database: process.env.DB_NAME,
  user: process.env.DB_USER,
  password: process.env.DB_PASSWORD,
});

// Database query tool
{
  name: "query-customers",
  description: "Query customer database",
  inputSchema: {
    type: "object",
    properties: {
      search: {
        type: "string",
        description: "Customer name or ID to search for",
      },
      limit: {
        type: "number",
        description: "Maximum number of results",
        default: 10,
      },
    },
    required: ["search"],
  },
}

// Handler
const result = await pool.query(
  "SELECT id, name, email FROM customers WHERE name ILIKE $1 LIMIT $2",
  [`%${search}%`, limit]
);

return {
  content: [
    {
      type: "text",
      text: JSON.stringify(result.rows, null, 2),
    },
  ],
};

Providing Resources

Resources are data sources AI can read:

import {
  ListResourcesRequestSchema,
  ReadResourceRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

// Enable resources capability
const server = new Server(
  {
    name: "docs-server",
    version: "1.0.0",
  },
  {
    capabilities: {
      resources: {},
    },
  }
);

// List available resources
server.setRequestHandler(ListResourcesRequestSchema, async () => {
  return {
    resources: [
      {
        uri: "docs://api-reference",
        name: "API Reference",
        description: "Complete API documentation",
        mimeType: "text/plain",
      },
      {
        uri: "docs://getting-started",
        name: "Getting Started Guide",
        description: "Quick start guide for new users",
        mimeType: "text/plain",
      },
    ],
  };
});

// Read resource content
server.setRequestHandler(ReadResourceRequestSchema, async (request) => {
  const uri = request.params.uri;

  if (uri === "docs://api-reference") {
    const content = await loadApiDocs();
    return {
      contents: [
        {
          uri,
          mimeType: "text/plain",
          text: content,
        },
      ],
    };
  }

  if (uri === "docs://getting-started") {
    const content = await loadGettingStarted();
    return {
      contents: [
        {
          uri,
          mimeType: "text/plain",
          text: content,
        },
      ],
    };
  }

  throw new Error(`Unknown resource: ${uri}`);
});

Creating Prompts

Prompts are reusable templates:

import {
  ListPromptsRequestSchema,
  GetPromptRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";

// Enable prompts capability
const server = new Server(
  {
    name: "prompt-server",
    version: "1.0.0",
  },
  {
    capabilities: {
      prompts: {},
    },
  }
);

// List available prompts
server.setRequestHandler(ListPromptsRequestSchema, async () => {
  return {
    prompts: [
      {
        name: "code-review",
        description: "Review code for quality and best practices",
        arguments: [
          {
            name: "language",
            description: "Programming language",
            required: true,
          },
          {
            name: "code",
            description: "Code to review",
            required: true,
          },
        ],
      },
      {
        name: "summarize-text",
        description: "Summarize long text into key points",
        arguments: [
          {
            name: "text",
            description: "Text to summarize",
            required: true,
          },
          {
            name: "length",
            description: "Desired summary length (short/medium/long)",
            required: false,
          },
        ],
      },
    ],
  };
});

// Get prompt content
server.setRequestHandler(GetPromptRequestSchema, async (request) => {
  if (request.params.name === "code-review") {
    const { language, code } = request.params.arguments as {
      language: string;
      code: string;
    };

    return {
      messages: [
        {
          role: "user",
          content: {
            type: "text",
            text: `Review this ${language} code for:
1. Code quality and readability
2. Best practices
3. Potential bugs
4. Security issues
5. Performance concerns

Code:
\`\`\`${language}
${code}
\`\`\`

Provide specific, actionable feedback.`,
          },
        },
      ],
    };
  }

  if (request.params.name === "summarize-text") {
    const { text, length = "medium" } = request.params.arguments as {
      text: string;
      length?: string;
    };

    const lengthGuide = {
      short: "3-5 bullet points",
      medium: "1-2 paragraphs",
      long: "detailed summary with sections",
    };

    return {
      messages: [
        {
          role: "user",
          content: {
            type: "text",
            text: `Summarize the following text into ${lengthGuide[length]}:

${text}`,
          },
        },
      ],
    };
  }

  throw new Error(`Unknown prompt: ${request.params.name}`);
});

Complete Example: File System Server

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
import {
  CallToolRequestSchema,
  ListToolsRequestSchema,
  ListResourcesRequestSchema,
  ReadResourceRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";
import * as fs from "fs/promises";
import * as path from "path";

class FileSystemServer {
  private server: Server;
  private rootPath: string;

  constructor(rootPath: string) {
    this.rootPath = rootPath;

    this.server = new Server(
      {
        name: "filesystem-server",
        version: "1.0.0",
      },
      {
        capabilities: {
          tools: {},
          resources: {},
        },
      }
    );

    this.setupHandlers();
  }

  private setupHandlers() {
    // List tools
    this.server.setRequestHandler(ListToolsRequestSchema, async () => {
      return {
        tools: [
          {
            name: "read-file",
            description: "Read contents of a file",
            inputSchema: {
              type: "object",
              properties: {
                path: { type: "string", description: "File path" },
              },
              required: ["path"],
            },
          },
          {
            name: "write-file",
            description: "Write contents to a file",
            inputSchema: {
              type: "object",
              properties: {
                path: { type: "string", description: "File path" },
                content: { type: "string", description: "File content" },
              },
              required: ["path", "content"],
            },
          },
          {
            name: "list-directory",
            description: "List files in a directory",
            inputSchema: {
              type: "object",
              properties: {
                path: { type: "string", description: "Directory path" },
              },
              required: ["path"],
            },
          },
        ],
      };
    });

    // Handle tool calls
    this.server.setRequestHandler(CallToolRequestSchema, async (request) => {
      const { name, arguments: args } = request.params;

      switch (name) {
        case "read-file":
          return this.readFile(args.path as string);
        case "write-file":
          return this.writeFile(args.path as string, args.content as string);
        case "list-directory":
          return this.listDirectory(args.path as string);
        default:
          throw new Error(`Unknown tool: ${name}`);
      }
    });

    // List resources
    this.server.setRequestHandler(ListResourcesRequestSchema, async () => {
      const files = await this.getAllFiles(this.rootPath);
      return {
        resources: files.map((filePath) => ({
          uri: `file://${filePath}`,
          name: path.basename(filePath),
          description: `File: ${path.relative(this.rootPath, filePath)}`,
          mimeType: "text/plain",
        })),
      };
    });

    // Read resource
    this.server.setRequestHandler(ReadResourceRequestSchema, async (request) => {
      const uri = request.params.uri;
      if (uri.startsWith("file://")) {
        const filePath = uri.slice(7);
        const content = await fs.readFile(filePath, "utf-8");
        return {
          contents: [
            {
              uri,
              mimeType: "text/plain",
              text: content,
            },
          ],
        };
      }
      throw new Error(`Unknown resource: ${uri}`);
    });
  }

  private async readFile(filePath: string) {
    const fullPath = path.join(this.rootPath, filePath);
    const content = await fs.readFile(fullPath, "utf-8");
    return {
      content: [{ type: "text", text: content }],
    };
  }

  private async writeFile(filePath: string, content: string) {
    const fullPath = path.join(this.rootPath, filePath);
    await fs.writeFile(fullPath, content, "utf-8");
    return {
      content: [{ type: "text", text: `File written: ${filePath}` }],
    };
  }

  private async listDirectory(dirPath: string) {
    const fullPath = path.join(this.rootPath, dirPath);
    const entries = await fs.readdir(fullPath, { withFileTypes: true });
    const list = entries.map((entry) => ({
      name: entry.name,
      type: entry.isDirectory() ? "directory" : "file",
    }));
    return {
      content: [{ type: "text", text: JSON.stringify(list, null, 2) }],
    };
  }

  private async getAllFiles(dir: string): Promise<string[]> {
    const entries = await fs.readdir(dir, { withFileTypes: true });
    const files = await Promise.all(
      entries.map((entry) => {
        const fullPath = path.join(dir, entry.name);
        return entry.isDirectory() ? this.getAllFiles(fullPath) : [fullPath];
      })
    );
    return files.flat();
  }

  async start() {
    const transport = new StdioServerTransport();
    await this.server.connect(transport);
    console.error("Filesystem MCP server running");
  }
}

// Run server
const server = new FileSystemServer(process.env.ROOT_PATH || "./");
server.start().catch(console.error);

Configuration

VS Code Configuration

Add your MCP server to VS Code settings for GitHub Copilot:

Option 1: VS Code settings.json

{
  "github.copilot.mcpServers": {
    "filesystem": {
      "command": "node",
      "args": ["/path/to/filesystem-server/build/index.js"],
      "env": {
        "ROOT_PATH": "/path/to/allowed/directory"
      }
    }
  }
}

Option 2: Workspace settings

Create .vscode/settings.json in your project:

{
  "github.copilot.mcpServers": {
    "project-tools": {
      "command": "node",
      "args": ["./tools/mcp-server.js"]
    }
  }
}

Best Practices

Error Handling

• Return meaningful error messages
• Use isError: true for error responses
• Log errors for debugging
• Validate inputs thoroughly

Security

• Validate and sanitize all inputs
• Implement proper authentication if needed
• Restrict file system access to specific directories
• Never expose sensitive data
• Rate limit expensive operations

Performance

• Cache frequently accessed data
• Use streaming for large datasets
• Implement pagination for lists
• Consider async operations

Testing

// Test tool handler
import { describe, it, expect } from "vitest";

describe("Weather Tool", () => {
  it("should return weather data", async () => {
    const result = await server.callTool("get-weather", {
      location: "San Francisco",
    });

    expect(result.content[0].type).toBe("text");
    expect(JSON.parse(result.content[0].text)).toHaveProperty("temperature");
  });
});

Deployment

npm Package

{
  "name": "my-mcp-server",
  "version": "1.0.0",
  "type": "module",
  "main": "build/index.js",
  "bin": {
    "my-mcp-server": "build/index.js"
  },
  "scripts": {
    "build": "tsc",
    "start": "node build/index.js"
  }
}

Docker Container

FROM node:18-alpine

WORKDIR /app

COPY package*.json ./
RUN npm ci --production

COPY build ./build

CMD ["node", "build/index.js"]

Resources

• MCP Specification
• MCP SDK Documentation
• Example MCP Servers
• GitHub Copilot Documentation
• VS Code Extension API

References

• Model Context Protocol
• MCP TypeScript SDK
• Building MCP Servers
• GitHub Copilot Extensibility