Module 4: Available MCP Servers

---

The MCP Server Ecosystem

One of MCP's greatest strengths is its growing ecosystem of pre-built servers. Instead of building integrations from scratch, you can often find a server that does exactly what you need. This module catalogs the most useful servers, explains what each one does, and helps you choose the right ones for your workflow.

---

Official Anthropic Servers

These servers are maintained by Anthropic and represent the most stable, well-documented options.

---

Filesystem Server

Package: @modelcontextprotocol/server-filesystem

Purpose: Read and search local files and directories.

Capabilities:

• List directory contents
• Read file contents
• Search files by name patterns
• Get file metadata

Configuration:

{
  "filesystem": {
    "command": "npx",
    "args": [
      "-y",
      "@modelcontextprotocol/server-filesystem",
      "/path/to/directory1",
      "/path/to/directory2"
    ]
  }
}

Use Cases:

• Code review and analysis
• Documentation lookup
• Configuration file management
• Log file analysis

Security Note: The server only accesses directories explicitly listed in args. It cannot escape these boundaries.

---

GitHub Server

Package: @modelcontextprotocol/server-github

Purpose: Interact with GitHub repositories, issues, and pull requests.

Capabilities:

• List and read repositories
• Create and manage issues
• Review pull requests
• Read file contents from repos
• Search code across repositories

Configuration:

{
  "github": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-github"],
    "env": {
      "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_xxxxxxxxxxxx"
    }
  }
}

Getting a Token:

1. Go to GitHub Settings → Developer Settings → Personal Access Tokens
2. Generate a new token (classic) with repo scope
3. Copy the token to your configuration

Use Cases:

• Automated code review
• Issue triage and management
• Repository analysis
• Documentation generation from code

---

GitLab Server

Package: @modelcontextprotocol/server-gitlab

Purpose: Similar to GitHub server but for GitLab repositories.

Capabilities:

• Repository management
• Merge request operations
• Issue tracking
• CI/CD pipeline access

Configuration:

{
  "gitlab": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-gitlab"],
    "env": {
      "GITLAB_PERSONAL_ACCESS_TOKEN": "glpat-xxxxxxxxxxxx",
      "GITLAB_API_URL": "https://gitlab.com/api/v4"
    }
  }
}

---

PostgreSQL Server

Package: @modelcontextprotocol/server-postgres

Purpose: Query and analyze PostgreSQL databases.

Capabilities:

• Execute SELECT queries
• Describe table schemas
• List tables and views
• Analyze query results

Configuration:

{
  "postgres": {
    "command": "npx",
    "args": [
      "-y",
      "@modelcontextprotocol/server-postgres",
      "postgresql://user:password@host:5432/database"
    ]
  }
}

Security Note: The server is read-only by default. It won't execute INSERT, UPDATE, DELETE, or DDL statements.

Use Cases:

• Data analysis and reporting
• Schema documentation
• Query optimization suggestions
• Data quality checks

---

SQLite Server

Package: @modelcontextprotocol/server-sqlite

Purpose: Query local SQLite database files.

Capabilities:

• Read-only database queries
• Schema inspection
• Table listing

Configuration:

{
  "sqlite": {
    "command": "npx",
    "args": [
      "-y",
      "@modelcontextprotocol/server-sqlite",
      "/path/to/database.db"
    ]
  }
}

Use Cases:

• Analyzing application databases
• Browser history/bookmark analysis
• Mobile app database inspection
• Local data exploration

---

Brave Search Server

Package: @modelcontextprotocol/server-brave-search

Purpose: Search the web using Brave Search API.

Capabilities:

• Web search queries
• Local search (businesses, places)
• News search

Configuration:

{
  "brave-search": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-brave-search"],
    "env": {
      "BRAVE_API_KEY": "BSA_xxxxxxxxxxxx"
    }
  }
}

Getting an API Key:

1. Sign up at brave.com/search/api
2. Choose a plan (free tier available)
3. Copy your API key

Use Cases:

• Research assistance
• Fact-checking
• Finding documentation
• Competitive analysis

---

Puppeteer Server

Package: @modelcontextprotocol/server-puppeteer

Purpose: Browser automation and web scraping.

Capabilities:

• Navigate to URLs
• Take screenshots
• Extract page content
• Execute JavaScript in pages
• Fill forms and click buttons

Configuration:

{
  "puppeteer": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-puppeteer"]
  }
}

Use Cases:

• Web scraping for research
• Automated testing
• Screenshot capture
• Form automation

Note: This server launches a real browser. Be cautious about which sites you automate.

---

Memory Server

Package: @modelcontextprotocol/server-memory

Purpose: Persistent memory storage for notes and context.

Capabilities:

• Store key-value pairs
• Retrieve stored information
• List all stored items
• Delete items

Configuration:

{
  "memory": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-memory"]
  }
}

Use Cases:

• Storing project notes
• Remembering preferences
• Building knowledge bases
• Cross-session context

---

Fetch Server

Package: @modelcontextprotocol/server-fetch

Purpose: Fetch and read web content.

Capabilities:

• GET requests to URLs
• Extract text content from HTML
• Handle redirects
• Read raw content or parsed text

Configuration:

{
  "fetch": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-fetch"]
  }
}

Use Cases:

• Reading documentation
• Fetching API responses
• Analyzing web pages
• Content summarization

---

Community Servers

Beyond official servers, the community has created many useful integrations.

---

Slack Server

Package: @modelcontextprotocol/server-slack

Purpose: Interact with Slack workspaces.

Capabilities:

• Read channel messages
• Post messages
• List channels
• Search messages

Configuration:

{
  "slack": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-slack"],
    "env": {
      "SLACK_BOT_TOKEN": "xoxb-xxxxxxxxxxxx",
      "SLACK_TEAM_ID": "T01234567"
    }
  }
}

---

Google Drive Server

Package: @modelcontextprotocol/server-gdrive

Purpose: Access files in Google Drive.

Capabilities:

• List files and folders
• Read file contents
• Search by name or content
• Access shared drives

Configuration:

{
  "google-drive": {
    "command": "npx",
    "args": ["-y", "@modelcontextprotocol/server-gdrive"],
    "env": {
      "GDRIVE_CREDENTIALS_PATH": "/path/to/credentials.json"
    }
  }
}

Note: Requires Google Cloud project and OAuth setup.

---

Notion Server

Purpose: Access Notion pages and databases.

Capabilities:

• Read pages
• Query databases
• Search content

Use Cases:

• Knowledge base access
• Project documentation
• Note retrieval

---

Linear Server

Purpose: Access Linear issue tracking.

Capabilities:

• Read issues
• Create issues
• Update status
• Query projects

Use Cases:

• Issue triage
• Sprint planning
• Status reporting

---

Choosing the Right Servers

When deciding which servers to set up, consider:

1. Start with your pain points What tasks do you find yourself doing repeatedly that Claude could help with if it had access?

2. Consider security implications Each server is an access point. Only enable what you actually need.

3. Check token/API requirements Some servers need API keys. Make sure you can obtain them before configuring.

4. Test in isolation Add one server at a time. Verify it works before adding more.

---

Server Capability Matrix

Server	Resources	Tools	Prompts	Auth Required
Filesystem	Yes	Yes	No	No
GitHub	Yes	Yes	No	Token
PostgreSQL	Yes	Yes	No	Connection string
SQLite	Yes	Yes	No	No
Brave Search	No	Yes	No	API Key
Puppeteer	No	Yes	No	No
Memory	Yes	Yes	No	No
Fetch	No	Yes	No	No
Slack	Yes	Yes	No	Token
Google Drive	Yes	Yes	No	OAuth

---

Practical Configuration Examples

Developer Setup:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "."]
    },
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": { "GITHUB_PERSONAL_ACCESS_TOKEN": "${GITHUB_TOKEN}" }
    }
  }
}

Researcher Setup:

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "~/Research"]
    },
    "brave-search": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-brave-search"],
      "env": { "BRAVE_API_KEY": "${BRAVE_API_KEY}" }
    },
    "fetch": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-fetch"]
    }
  }
}

Data Analyst Setup:

{
  "mcpServers": {
    "postgres": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-postgres", "${DATABASE_URL}"]
    },
    "filesystem": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-filesystem", "~/data", "~/reports"]
    }
  }
}

---

Finding More Servers

The MCP ecosystem is growing. To find more servers:

Official Registry: Check the MCP GitHub organization for official and featured servers.

NPM Search: Search npm for @modelcontextprotocol to find published servers.

Community Resources: Look for awesome-mcp lists and community curations.

Build Your Own: If you can't find what you need, you can build it. We'll cover that in Module 6.

---

Key Takeaways

1. Rich ecosystem - Dozens of pre-built servers for common integrations

2. Official servers - Filesystem, GitHub, databases, search, and more

3. Community servers - Slack, Notion, Linear, and growing

4. Configuration patterns - Most servers use similar npx-based setup

5. Security first - Only enable servers you need, with minimal permissions

6. Check requirements - Many servers need API keys or tokens

7. Start small - Add servers incrementally, testing each one

---

Looking Ahead

Now that you know what servers are available, the next module focuses on using MCP specifically with Claude Code. You'll learn project-scoped configurations, development workflows, and how to maximize MCP in your coding workflow.

---

Next up: Module 5 - Using MCP with Claude Code