MCP Servers: A Practical Getting Started Guide
MCP servers let you give AI access to your data and tools. In this guide, we’ll connect your first server in 15 minutes.
What You Need
- Claude Desktop (free)
- Node.js 18+ installed
- 15 minutes
Step 1: Install Claude Desktop
Download Claude Desktop from claude.ai and install it. It’s free.
Step 2: Configure the Filesystem MCP Server
Open the Claude Desktop configuration:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Add the server configuration:
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-filesystem",
"/Users/yourname/Documents"
]
}
}
}
Replace /Users/yourname/Documents with your actual folder.
Step 3: Restart Claude Desktop
Close and reopen Claude Desktop. You’ll see a hammer icon in the interface — this means MCP tools are available.
Step 4: Test It
Try typing in chat:
Show me the files in my Documents folder
Claude will use the MCP server to read the filesystem and display the results.
More Servers You Can Connect
GitHub MCP Server
{
"mcpServers": {
"github": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-github"],
"env": {
"GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
}
}
}
}
Now Claude can:
- Browse your repositories
- Create issues and pull requests
- Search code
PostgreSQL MCP Server
{
"mcpServers": {
"postgres": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-postgres",
"postgresql://user:pass@localhost:5432/mydb"
]
}
}
}
Claude gets access to your database and can:
- Execute SQL queries
- Analyze data
- Generate reports
Security Best Practices
- Limit access — only grant access to folders/databases you need
- Read-only where possible — if AI only reads, don’t give write permissions
- Don’t store secrets in config — use environment variables
- Review actions — Claude always asks permission before executing
Next Steps
Ready to build your own MCP server? Read the next guide in our documentation.
AI Automation Researcher. Researches AI for corporate AI automation — agents, tools, and prompt engineering.
Related articles
OpenAI Assistants API: The Complete Guide
How to build an AI assistant with OpenAI Assistants API — function calling, file search, code interpreter.
What Are AI Agents and Why Your Business Needs Them
Understanding AI agents — how they work, what makes them different from chatbots, and why they matter.
How AI Agents Are Transforming Business in 2026: Real Case Studies
5 real examples of how companies use AI agents for automation, support, and analytics.
Stay updated on AI
Get weekly insights on AI agents, tools, and prompt engineering delivered to your inbox.