# Design Database Schemas With DB Craft

Database schema design can become fragmented across notes, SQL drafts, and diagrams. DB Craft opens a local visual studio to model tables, validate structure, and export SQL.

## Install

```bash
npx skillstore add laozhu001/dbcraft
```

## Metadata

- Status: approved
- Slug: laozhu001-dbcraft
- Skillstore revision: r1
- Version status: missing
- Tree hash: 56c409f3752fe39ebcc98c12a6f34a0016ce3c8f8b6bdd3f1679cc0a806007b8
- Author: laozhu001
- GitHub username: laozhu001
- License: MIT
- Repository: https://github.com/laozhu001/dbcraft/tree/dev-laozhu/skill
- Ref: a25199bc7d6b82598536822d1738eb5d5f54025b
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: confirmation\_required
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, env\_access, network
- Quality score: 38
- Quality tier: warning
- Public page: https://skillstore.pages.dev/skills/laozhu001-dbcraft
- Manifest: https://skillstore.pages.dev/api/skills/laozhu001-dbcraft/manifest

## Capabilities

- Opens the local DB Craft visual schema studio in a browser.
- Creates and edits models for MySQL 8, PostgreSQL 14, SQLite, and MSSQL.
- Imports existing CREATE TABLE scripts into the visual model.
- Generates table designs with a configured AI provider or Codex handoff.
- Saves DB Craft model files and exports SQL into the active workspace.
- Guides validation, auto-layout, and handoff to migration workflows.

## Use Cases

- Plan New Service Tables: Create a visual model for a new service, validate table structure, and export SQL for implementation review.
- Convert Requirements Into Models: Turn product requirements or screenshots into tables that align frontend data needs with backend storage.
- Review Schema Handoffs: Inspect relationships, generated tables, and exported SQL before a separate migration workflow applies changes.

## Prompt Templates

### Open A New Model

```
Open DB Craft and create a MySQL 8 model for a student management system. Add student, class, and teacher tables.
```

### Import Existing SQL

```
Open DB Craft and import my existing table definitions into the current workspace model. Validate the structure and organize the diagram.
```

### Generate Tables With AI

```
Use DB Craft AI to design tables for an e-commerce cart, order, and payment workflow. Save the model in the active workspace.
```

### Prepare Migration Handoff

```
Continue from the current DB Craft handoff, update the model with the requested changes, validate it, export SQL, and summarize deployment notes.
```

## Limitations

- The current public package supports Windows with local Node.js and PowerShell.
- It does not execute schema changes against live databases by default.
- AI table generation may send prompts to the configured model provider.
- Browser directory permissions may require a manual user selection.

## Best Practices

- Keep each project model saved in the active workspace directory.
- Review generated SQL before using it in a real database workflow.
- Use personal API keys carefully and hide secrets from screenshots.

## Anti Patterns

- Treating DB Craft output as an automatic production database change.
- Starting duplicate local services before checking the existing service.
- Placing secrets in schema comments, AI prompts, or public screenshots.

## Security Audit

- Audited at: 2026-07-05T18:31:40.987\+00:00
- Summary: Most static hits are false positives from Markdown code spans, loopback URLs, local model filenames, and safety guidance. Confirmed risks remain: the skill can start a local DB Craft service through PowerShell and Node, and AI generation can use OPENAI\_API\_KEY. I found no prompt injection language or evidence of intentional data exfiltration.

## Stats

- Views: 122
- Downloads: 9
- Favorites: 0
- Popularity score: 0