Database Development Workflow with Prisma MCP

What You'll Build

A complete database development workflow that uses Continue CLI with Prisma MCP to design schemas, generate migrations, seed data, and manage your PostgreSQL database - all through natural language prompts.

What You’ll Learn

This cookbook teaches you to:

Use Prisma Remote MCP to build database workflows
Design database schemas with AI assistance
Generate and apply migrations automatically
Create realistic seed data for development
Launch Prisma Studio for visual database management
Set up automated database health checks

Prerequisites

Before starting, ensure you have:

Node.js 18+ installed locally
Continue CLI with active credits (required for API usage)
Basic understanding of SQL and database concepts
A project directory where you want to set up your database

Install Continue CLI

npm i -g @continuedev/cli

Set up Continue CLI Account & API Key

Visit Continue Organizations
Sign up or log in to your Continue account
Navigate to your organization settings
Click “API Keys” and then ”+ New API Key”
Copy the API key immediately (you won’t see it again!)
Login to the CLI: cn login

Add Prisma Secrets to Continue Hub

For the Prisma Remote MCP to work, you need to authenticate with Prisma:

Visit Continue Hub Secrets
Click ”+ Add Secret”
Add your Prisma authentication token:
- Key: prisma/prisma-remote-mcp/PRISMA_AUTH_TOKEN
- Value: Your Prisma auth token (get from Prisma Console)

The Prisma Remote MCP runs on Prisma’s hosted infrastructure and requires authentication to access your databases and projects.

Continue CLI with Prisma Remote MCP can generate complete database schemas, write migrations, and even create realistic seed data - you just describe what you need!

Step 1: Set Up Your Database

Prisma Postgres Setup
Local PostgreSQL

Prisma MCP works with any PostgreSQL database, including Prisma Postgres:

Create a Prisma Account (optional, for hosted databases)
- Visit Prisma Console
- Sign up or log in
Authenticate via Continue CLI Prompt:
```
Log me into the Prisma Console
```
The AI will use the Prisma-Login tool to authenticate.

Create a Database Prompt:

Create a new Prisma Postgres database in the US region for my e-commerce project

You can also use any PostgreSQL database by configuring your connection string in .env:

DATABASE_URL="postgresql://user:password@localhost:5432/mydb"

Prisma Database Workflow Options

Fastest Path to Success

Skip the manual setup and use our pre-built Prisma Continuous AI agent that includes optimized prompts, rules, and the Prisma MCP for consistent database development.

How Prisma Remote MCP Works:

Connects to Prisma’s hosted MCP server via mcp-remote
Runs on Prisma’s infrastructure (no local installation required)
Provides tools for migrations, schema management, and database operations
Integrates with Prisma Studio for visual database exploration
Includes AI safety guardrails to prevent accidental data loss

Available Prisma Remote MCP Tools:

migrate-status - Check migration status
migrate-dev - Create and execute migrations
migrate-reset - Reset the database
Prisma-Postgres-account-status - Check authentication
Create-Prisma-Postgres-Database - Provision new databases
Prisma-Login - Authenticate with Prisma
Prisma-Studio - Launch the data explorer

Installation: Add Prisma Remote MCP from Continue Hub

⚡ Quick Start (Recommended)
🛠️ Manual Setup

Perfect for: Rapid database development with AI-powered schema design and migration generation

Add the Pre-Built Agent

Visit the Prisma Continuous AI Agent on Continue Hub and click “Install Agent” or run:

cn --agent "continuedev/prisma-continuous-ai"

This agent includes:

Optimized prompts for schema design and migration workflows
Built-in rules for database best practices
Prisma Remote MCP for reliable database operations
Automatic schema validation and safety checks

Run Schema Design

Navigate to your project directory and enter this prompt in the Continue CLI TUI:

Design a database schema for an e-commerce platform with products, users, orders, and reviews

That’s it! The agent handles everything automatically.

Why Use the Agent? Get consistent results with pre-tested prompts and built-in database design best practices.

Agent Requirements

To use the pre-built agent, you need either:

Continue CLI Pro Plan with the models add-on, OR
Your own API keys added to Continue Hub secrets

The agent will automatically detect and use your configuration. For Prisma Remote MCP:

Prisma authentication token added to Continue Hub secrets
PostgreSQL database (local or hosted via Prisma Postgres)
Prisma Remote MCP connects to Prisma’s hosted infrastructure (no local installation required)

AI Safety Guardrails: Prisma detects when destructive commands like prisma migrate reset --force are invoked through AI agents and requires explicit user consent before proceeding.

Step 2: Design Your Database Schema with AI

Use Continue CLI to design your database schema through natural language prompts.

E-commerce Schema
SaaS Application Schema
Blog Platform Schema

Prompt:

Design a complete e-commerce database schema with:

Tables:
- Users (authentication, profiles, addresses)
- Products (variants, inventory, categories)
- Orders (line items, status tracking, shipping)
- Reviews (ratings, comments, helpful votes)
- Carts (temporary shopping carts)

Requirements:
- Use UUIDs for all IDs
- Add proper relationships (foreign keys)
- Include created_at and updated_at timestamps
- Add soft deletes for users and products
- Create indexes for:
  * Product searches (name, category)
  * Order lookups (user_id, status, created_at)
  * Review queries (product_id, user_id)

Generate the Prisma schema and initial migration.

Prisma Remote MCP Capabilities for Schema Design:

Validates schema syntax automatically
Suggests optimal data types and relationships
Recommends indexes based on query patterns
Warns about potential performance issues
Leverages Prisma’s hosted AI models for enhanced schema recommendations

Step 3: Generate and Apply Migrations

Create database migrations from your schema design and apply them safely. Example: Complete Migration Workflow

I've designed my Prisma schema. Now I need to:

1. Check the current migration status
2. Create a new migration named "initial_schema"
3. Review the generated SQL before applying
4. Apply the migration to the database
5. Verify all tables were created correctly

For the migration:
- Include all schema changes
- Add helpful comments in the SQL
- Show me the migration file contents
- Confirm the migration was applied successfully

Database: PostgreSQL
Environment: Development

Expected Output: The AI will:

Run prisma migrate status to check current state
Generate migration with prisma migrate dev --name initial_schema
Show you the SQL file contents for review
Apply the migration automatically
Verify tables exist using database inspection

Best Practice: Always review AI-generated migrations before applying to production. Test in development first!

Step 4: Seed Your Database with Realistic Data

Generate realistic seed data for development and testing.

E-commerce Seed Data
SaaS Seed Data

Prompt:

Create a comprehensive seed script for my e-commerce database:

Generate:
- 50 users with realistic names and emails
- 200 products across 10 categories
  * Include product variants (size, color)
  * Set realistic prices and inventory
- 100 orders with random items
  * Various order statuses (pending, shipped, delivered)
  * Created at different dates over the past 6 months
- 300 product reviews
  * Ratings from 1-5 stars
  * Realistic review text
  * Mix of helpful/not helpful votes

Requirements:
- Use realistic fake data (not just "User 1", "Product 1")
- Maintain referential integrity
- Distribute data realistically (more delivered than pending orders)
- Add some edge cases (cancelled orders, out-of-stock items)

Create the seed.ts file and show me how to run it.

Prisma Seed Best Practices:

Use faker or similar libraries for realistic data
Run seeds in transactions for consistency
Make seeds idempotent (safe to run multiple times)
Include edge cases and test scenarios

Step 5: Launch Prisma Studio for Visual Management

Use Prisma Studio to visually explore and manage your database. Prompt:

Launch Prisma Studio so I can visually inspect my database tables and data

The AI will:

Run prisma studio command
Open Studio in your browser (usually http://localhost:5555)
Show you how to navigate the interface

Prisma Studio Features:

Browse all tables and relationships
Edit records directly
Filter and search data
View relationship connections visually

Step 6: Automate Database Health Checks

Set up automated database monitoring with Continue CLI and GitHub Actions:

name: Database Health Monitor

on:
  push:
    branches:
      - main
    paths:
      - 'prisma/**'
  pull_request:
    paths:
      - 'prisma/**'
  schedule:
    # Run daily at 3 AM UTC
    - cron: "0 3 * * *"
  workflow_dispatch: # Allow manual triggers

jobs:
  check-database-health:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - uses: actions/setup-node@v4
        with:
          node-version: "20"

      - name: Install Dependencies
        run: |
          npm install
          npm install -g @continuedev/cli
          echo "✅ Dependencies installed"

      - name: Setup Database
        env:
          DATABASE_URL: ${{ secrets.DATABASE_URL }}
        run: |
          npx prisma migrate deploy
          echo "✅ Migrations applied"

      - name: Analyze Database Schema
        env:
          CONTINUE_API_KEY: ${{ secrets.CONTINUE_API_KEY }}
          DATABASE_URL: ${{ secrets.DATABASE_URL }}
        run: |
          echo "🔍 Analyzing database health..."

          cn -p "Using Prisma MCP, perform a comprehensive database health check:

                 1. Migration Status:
                    - Check if all migrations are applied
                    - Identify pending migrations
                    - Detect migration conflicts

                 2. Schema Analysis:
                    - Verify all foreign key relationships are valid
                    - Check for missing indexes on foreign keys
                    - Identify tables without primary keys
                    - Find unused indexes (if possible)

                 3. Data Integrity:
                    - Check for orphaned records (broken foreign keys)
                    - Verify required fields don't have nulls
                    - Identify duplicate records in unique columns

                 4. Performance Review:
                    - Suggest missing indexes based on schema
                    - Identify large tables that might need partitioning
                    - Check for inefficient data types

                 5. Best Practices:
                    - Verify created_at/updated_at timestamps exist
                    - Check for soft delete implementations
                    - Ensure proper enum usage

                 Generate a detailed health report with:
                 - Critical issues that need immediate attention
                 - Warnings for potential problems
                 - Optimization suggestions
                 - SQL scripts to fix identified issues

                 If critical issues are found, fail the workflow."

      - name: Save Health Report
        if: always()
        run: |
          echo "## 📊 Database Health Report" >> $GITHUB_STEP_SUMMARY
          echo "Generated at $(date)" >> $GITHUB_STEP_SUMMARY
          echo "" >> $GITHUB_STEP_SUMMARY
          echo "Check the 'Analyze Database Schema' step for the full report" >> $GITHUB_STEP_SUMMARY

      - name: Upload Health Report
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: database-health-report
          path: |
            *.md
            *.sql

Required GitHub Secrets:

CONTINUE_API_KEY: Your Continue API key from hub.continue.dev/settings/api-keys
DATABASE_URL: PostgreSQL connection string for your test database

Add these at: Repository Settings → Secrets and variables → Actions

What You’ve Built

After completing this guide, you have a complete AI-powered database development workflow that:

Designs schemas - AI generates complete database schemas from natural language descriptions
Creates migrations - Automatically generates and applies database migrations
Seeds data - Generates realistic test data for development
Monitors health - Runs automated checks for schema issues and performance problems
Visual management - Provides GUI access via Prisma Studio

Continuous AI Database Development

Your system now operates at Level 2 Continuous AI - AI handles schema design and migration generation with human oversight for production deployments.

Advanced Database Development Prompts

Enhance your workflow with these advanced Continue CLI prompts:

Schema Evolution

Analyze existing schema and suggest improvements for scalability, including partitioning strategies and denormalization opportunities

Query Optimization

Review slow queries and generate optimized indexes, explain plans, and query rewrites

Data Migration

Generate data transformation scripts when changing schema structure (e.g., splitting tables, normalizing data)

Backup & Recovery

Create backup strategies and generate recovery scripts for disaster scenarios

Real-World Use Cases

Use Case 1: Adding Multi-Tenancy to Existing Schema

Prompt:

I need to add multi-tenancy to my existing database schema:

Current tables: users, products, orders

Requirements:
- Add an organizations table
- Link all existing tables to organizations
- Ensure data isolation between tenants
- Migrate existing data to a default organization
- Add organization_id to all relevant indexes

Generate:
1. Updated Prisma schema with multi-tenant support
2. Migration to add organization_id columns
3. Data migration script to populate organization_id
4. Updated seed script with multi-tenant data
5. Queries demonstrating tenant isolation

Use Case 2: Adding Full-Text Search

Prompt:

Add full-text search to my blog platform:

Tables: posts (title, content), comments (text)

Requirements:
- Use PostgreSQL full-text search capabilities
- Create search indexes for posts and comments
- Add search ranking/relevance
- Support phrase matching and stemming
- Highlight matching terms in results

Generate:
1. Migration to add tsvector columns and GIN indexes
2. Prisma schema updates
3. Example search queries using raw SQL
4. Helper functions for search in TypeScript

Use Case 3: Implementing Audit Logging

Prompt:

Add comprehensive audit logging to track all data changes:

Tables to audit: users, orders, products

Requirements:
- Record all INSERT, UPDATE, DELETE operations
- Store: who made changes, when, what changed, old/new values
- Implement using PostgreSQL triggers or application-level
- Query audit history for specific records
- Implement data retention policy (keep 90 days)

Generate:
1. Audit log table schema
2. Triggers or Prisma middleware for logging
3. Queries to view audit history
4. Cleanup script for old audit records

Security Best Practices

Database Security Guidelines:

Never commit .env files with database credentials
Use environment variables for all sensitive configuration
Implement row-level security (RLS) for multi-tenant apps
Review all AI-generated SQL before running in production
Use read replicas for analytics queries
Enable SSL/TLS for database connections
Regularly rotate database passwords
Use least-privilege database users for applications

Troubleshooting

Prisma Remote MCP Connection Issues

If you encounter issues with Prisma Remote MCP:

Verify your PRISMA_AUTH_TOKEN is correctly set in Continue Hub Secrets
Check that the secret key is exactly: prisma/prisma-remote-mcp/PRISMA_AUTH_TOKEN
Ensure you have network connectivity to https://mcp.prisma.io
Verify your Prisma account has access to the database at Prisma Console
Check your DATABASE_URL is correctly set in .env (for local Prisma CLI operations)

Common Migration Issues

Issue	Solution
Migration conflicts	Reset development database with `prisma migrate reset`
Schema drift detected	Run `prisma migrate dev` to sync schema with database
Connection refused	Verify DATABASE_URL and database is running
Permission denied	Check database user has CREATE/ALTER privileges

AI Safety Guardrails

When Prisma Remote MCP detects destructive operations (like migrate reset):

You’ll see a confirmation prompt
Review what will be deleted
Type “yes” to confirm or “no” to cancel
The AI will wait for your response before proceeding

This safety feature prevents accidental data loss when AI agents run database commands. The Prisma Remote MCP enforces these guardrails on the server side for enhanced security.

Next Steps

Explore Prisma Client for type-safe database queries
Set up Prisma Accelerate for connection pooling
Configure Prisma Pulse for real-time database events
Learn about Prisma Optimize for query performance insights
Join the Continue Discord for support

Guides

Cookbooks

Database Development Workflow with Prisma MCP

What You'll Build

What You’ll Learn

Prerequisites

Step 1: Set Up Your Database

Prisma Database Workflow Options

Fastest Path to Success

Step 2: Design Your Database Schema with AI

Step 3: Generate and Apply Migrations

Step 4: Seed Your Database with Realistic Data

Step 5: Launch Prisma Studio for Visual Management

Step 6: Automate Database Health Checks

What You’ve Built

Continuous AI Database Development

Advanced Database Development Prompts

Schema Evolution

Query Optimization

Data Migration

Backup & Recovery

Real-World Use Cases

Use Case 1: Adding Multi-Tenancy to Existing Schema

Use Case 2: Adding Full-Text Search

Use Case 3: Implementing Audit Logging

Security Best Practices

Troubleshooting

Prisma Remote MCP Connection Issues

Common Migration Issues

AI Safety Guardrails

Next Steps

Resources

Guides

Cookbooks

What You'll Build

​What You’ll Learn

​Prerequisites

​Step 1: Set Up Your Database

​Prisma Database Workflow Options

Fastest Path to Success

​Step 2: Design Your Database Schema with AI

​Step 3: Generate and Apply Migrations

​Step 4: Seed Your Database with Realistic Data

​Step 5: Launch Prisma Studio for Visual Management

​Step 6: Automate Database Health Checks

​What You’ve Built

Continuous AI Database Development

​Advanced Database Development Prompts

Schema Evolution

Query Optimization

Data Migration

Backup & Recovery

​Real-World Use Cases

​Use Case 1: Adding Multi-Tenancy to Existing Schema

​Use Case 2: Adding Full-Text Search

​Use Case 3: Implementing Audit Logging

​Security Best Practices

​Troubleshooting

​Prisma Remote MCP Connection Issues

​Common Migration Issues

​AI Safety Guardrails

​Next Steps

​Resources

What You’ll Learn

Prerequisites

Step 1: Set Up Your Database

Prisma Database Workflow Options

Step 2: Design Your Database Schema with AI

Step 3: Generate and Apply Migrations

Step 4: Seed Your Database with Realistic Data

Step 5: Launch Prisma Studio for Visual Management

Step 6: Automate Database Health Checks

What You’ve Built

Advanced Database Development Prompts

Real-World Use Cases

Use Case 1: Adding Multi-Tenancy to Existing Schema

Use Case 2: Adding Full-Text Search

Use Case 3: Implementing Audit Logging

Security Best Practices

Troubleshooting

Prisma Remote MCP Connection Issues

Common Migration Issues

AI Safety Guardrails

Next Steps

Resources