Glama — MCP Server Discovery and Management

What it is

Glama is a discovery and management platform for Model Context Protocol (MCP) servers. It maintains a curated directory of MCP servers that extend AI coding tools like Claude Code and Cline with capabilities such as database access, file system operations, and API integrations.

The platform is aimed at developers who use AI coding assistants and want to find MCP servers by category, read install instructions, and monitor server health. Instead of searching GitHub for individual MCP server repos, Glama aggregates them into a single browsable catalog.

How it saves time or tokens

Finding the right MCP server for a specific integration typically involves searching GitHub, reading multiple READMEs, and comparing configuration formats. Glama consolidates this process into a single search. Each server listing includes standardized install instructions, configuration snippets, and compatibility information.

For teams managing multiple MCP servers, Glama provides monitoring and management features that reduce the overhead of keeping server configurations in sync across developer machines.

How to use

1. Visit the Glama directory at glama.ai/mcp/servers
2. Browse MCP servers by category or search by name
3. Click any server to view install instructions and configuration
4. Copy the configuration to your .mcp.json or Claude Code settings

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

Example

# Search for a PostgreSQL MCP server on Glama
# Then add the config to your project:
cat > .mcp.json << 'EOF'
{
  "mcpServers": {
    "postgres": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-postgres"],
      "env": {
        "DATABASE_URL": "postgresql://user:pass@localhost:5432/mydb"
      }
    }
  }
}
EOF

After adding the configuration, your AI coding tool connects to the MCP server and gains access to the database for schema inspection, query execution, and data exploration.

Related on TokRepo

• MCP integrations -- Browse specific MCP server configurations for GitHub and other services
• AI coding tools -- Explore tools that enhance your AI-assisted development workflow

Common pitfalls

• MCP server configurations vary between AI tools. A config that works in Claude Code may need adjustments for Cline or other clients. Always check the tool-specific instructions on Glama.
• Some MCP servers require environment variables with credentials. Never commit these to version control. Use .env files or secret managers instead.
• MCP servers run as local processes. Installing many servers simultaneously increases memory usage. Start with the servers you actively need and add more as required.