Skip to main content
Glama

!tables

View and manage database tables across PostgreSQL, MySQL, and Firestore with this MCP tool, enabling efficient exploration and organization of structured data.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • src/index.ts:125-154 (registration)
    Registration of the '!tables' MCP tool, including the inline handler function that checks database connection and retrieves tables via dbService.getTables(), returning JSON or error.
    server.tool( "!tables", {}, async () => { if (!dbService) { return { content: [{ type: "text", text: "You must connect to a database first!" }], isError: true, }; } try { const tables = await dbService.getTables(); return { content: [ { type: "text", text: JSON.stringify(tables, null, 2), }, ], }; } catch (error: unknown) { const errorMessage = error instanceof Error ? error.message : 'Unknown error'; return { content: [{ type: "text", text: `Failed to get table information: ${errorMessage}` }], isError: true, }; } } );
  • The core handler logic in DatabaseService.getTables() that queries the database-specific information_schema or equivalent to fetch table names and column metadata (name, type, nullable). Supports PostgreSQL, MySQL, and Firestore.
    async getTables(): Promise<TableInfo[]> { switch (this.config.type) { case 'postgres': { const query = ` SELECT table_name, json_agg(json_build_object( 'name', column_name, 'type', data_type, 'nullable', is_nullable = 'YES' )) as columns FROM information_schema.columns WHERE table_schema = 'public' GROUP BY table_name; `; const result = await this.postgresClient!.query(query); return result.rows.map(row => ({ name: row.table_name, columns: row.columns })); } case 'mysql': { const [rows] = await this.mysqlConnection!.query(` SELECT TABLE_NAME as tableName, GROUP_CONCAT( JSON_OBJECT( 'name', COLUMN_NAME, 'type', DATA_TYPE, 'nullable', IS_NULLABLE = 'YES' ) ) as columns FROM information_schema.columns WHERE table_schema = DATABASE() GROUP BY TABLE_NAME; `); return rows.map((row: any) => ({ name: row.tableName, columns: JSON.parse(`[${row.columns}]`) })); } case 'firestore': { // Firestore'da tablo yapısı olmadığı için koleksiyonları listeleyeceğiz const collections = await this.firestoreClient!.listCollections(); return collections.map(collection => ({ name: collection.id, columns: [] // Firestore şemasız olduğu için boş bırakıyoruz })); } default: return []; } }
  • TypeScript interface definitions for TableInfo (table name and columns array) and ColumnInfo (column name, type, nullable flag), which structure the output data from !tables tool.
    export interface TableInfo { name: string; columns: ColumnInfo[]; } export interface ColumnInfo { name: string; type: string; nullable: boolean; }

Other Tools

Related Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/wirdes/db-mcp-tool'

If you have feedback or need assistance with the MCP directory API, please join our Discord server