MCP Scenario: Query Database and Save Results to JSON

Scenario Overview

This scenario demonstrates how to combine MCP PostgreSQL tool and Filesystem tool to complete the workflow from database query to result saving [^1].

Prerequisites

1. mcp-server-postgres configured
2. mcp-server-filesystem configured
3. Database accessible with query permissions

Complete Workflow

Step 1: Query Database

Use postgres/query tool to execute SQL query:

{
  "connection_string": "postgresql://user:pass@localhost:5432/mydb",
  "query": "SELECT * FROM users WHERE created_at > '2024-01-01'"
}

Step 2: Process Query Results

Query results returned in JSON format:

{
  "rows": [
    {"id": 1, "name": "Alice", "email": "alice@example.com"},
    {"id": 2, "name": "Bob", "email": "bob@example.com"}
  ],
  "rowCount": 2
}

Step 3: Save as JSON File

Use filesystem/write_file tool to save results:

{
  "path": "/Users/username/data/users_2024.json",
  "content": "{\"users\": [{\"id\": 1, ...}]}"
}

Error Handling

Database Connection Failure

• Check connection string format
• Confirm database service running
• Verify network connectivity

File Write Failure

• Check directory permissions
• Confirm path within allowed directories
• Verify disk space

Best Practices

1. Use parameterized queries to prevent SQL injection
2. Process large result sets in batches
3. Add timestamps to filenames to avoid overwriting
4. Validate JSON format before saving

Reference Sources

[^1]: MCP with Postgres - PostgreSQL MCP tool usage guide
[^2]: PostgreSQL MCP Server Setup - PostgreSQL MCP server configuration tutorial