data-modeling

Data Modeling

Purpose

Design robust, maintainable data models for relational and graph data stores with clear schema design principles, normalization strategies, and traversal patterns.

Agent Protocol

Trigger

Exact user phrases: "data model", "schema design", "normalization", "denormalization", "3NF", "table inheritance", "soft delete", "surrogate key", "graph model", "property graph", "RDF", "knowledge graph", "node edge model", "graph traversal".

Input Context

• Data store type (relational, graph, or hybrid)
• Access patterns (OLTP, OLAP, graph queries)
• Volume and growth expectations
• Consistency and integrity requirements
• Existing schema constraints and migration path
• Team expertise with relational vs graph technologies

Output Artifact

DDL statements, graph schema definitions, migration scripts. No file unless requested.

Response Format

## Entity: {name}
| Field | Type | Constraints | Notes |
|-------|------|-------------|-------|
| {field} | {type} | {constraints} | {notes} |

## Graph: {name}
Nodes: {node types with properties}
Edges: {edge types with properties}
Indexes: {indexed properties}
Traversal: {common query patterns}

No preamble. No postamble. No explanations. No filler/hedging/transitions.

Completion Criteria

• Entities normalized to appropriate normal form
• Table inheritance pattern selected (if applicable)
• Temporal tracking strategy defined (if needed)
• Soft delete vs hard delete decided per entity
• Surrogate vs composite key decision documented per table
• Graph node/edge schema defined with property types
• Graph traversal patterns identified and indexed
• Migration strategy defined (expand-contract or in-place)

Max Response Length

200 lines of schema and code.

Workflow

Step 1: Conceptual Model

Identify entities, relationships, and business rules independent of technology. Each entity represents a real-world object or concept. Relationships describe how entities interact. Business rules become constraints and invariants.