data-model-creation
[Deprecated] Optional advanced tool for complex data modeling. For simple MySQL table creation, use relational-database-tool directly; for PostgreSQL / CloudBase PG schema work, use postgresql-development. New environments should use PostgreSQL DDL via queryPgDatabase/managePgDatabase — see postgresql-development skill instead.
TencentCloudBase · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add TencentCloudBase/cloudbase-skills --skill data-model-creation技能说明
name: data-model-creation description: "[Deprecated] Optional advanced tool for complex data modeling. For simple MySQL table creation, use relational-database-tool directly; for PostgreSQL / CloudBase PG schema work, use postgresql-development. New environments should use PostgreSQL DDL via queryPgDatabase/managePgDatabase — see postgresql-development skill instead." version: 2.33.2 alwaysApply: false metadata: priority: "5" deprecated: "true"
Sibling skills (local only)
Sibling CloudBase skills ship beside this skill. Use local relative paths such as ../auth-tool-cloudbase/SKILL.md.
If a referenced sibling skill file is missing from this environment, ask the user to install the full CloudBase plugin (or the missing skill). Do not HTTP-fetch remote skill or protocol markdown into the agent context.
Data Model Creation
Activation Contract
Use this first when
- The user explicitly wants Mermaid
classDiagrammodeling. - The task needs complex multi-entity relational design, visual ER-style output, or generated data-model structure rather than direct SQL.
- You need to create CloudBase data models through the dedicated modeling tools, or you need to inspect an existing model before planning follow-up changes.
Read before writing code if
- The request mentions data model, ER diagram, Mermaid, relationship graph, or enterprise schema design.
- The user wants to reuse or update an existing published model.
Then also read
- Direct MySQL SQL creation or schema change ->
../relational-database-mcp-cloudbase/SKILL.md - PostgreSQL / CloudBase PG schema work ->
../postgresql-development-cloudbase/SKILL.md - Broader feature planning before schema work ->
../spec-workflow/SKILL.md
Do NOT use for
- Simple
CREATE TABLE,ALTER TABLE, or CRUD tasks. - Document-database collection design.
- Frontend-only data-shape discussions with no modeling requirement.
Common mistakes / gotchas
- Using Mermaid modeling for a task that only needs one or two SQL statements.
- Mixing SQL-table design and NoSQL collection design in the same model.
- Generating diagrams without first deciding entity boundaries and ownership relations.
- Publishing a new model before validating the generated fields and relationships.
Minimal checklist
- Confirm Mermaid modeling is actually needed.
- List the core entities and relationships first.
- Decide whether this is a new model or an update.
- Keep the initial model small unless the user explicitly wants a large enterprise schema.
Overview
This skill is an advanced modeling path, not the default path for database work.
- For most MySQL database tasks, use
relational-database-mcp-cloudbaseand write SQL directly. If the task says PostgreSQL, CloudBase PG, PG mode,app.rdb(),queryPgDatabase,managePgDatabase, or RLS, usepostgresql-development-cloudbaseinstead. - Use this skill only when diagram-driven modeling adds value.
Quick routing
Use relational-database-mcp-cloudbase instead when
- You need MySQL
CREATE TABLE,ALTER TABLE,INSERT,UPDATE,DELETE, orSELECT - The schema is small and already clear
- The user never asked for a visual model
- The task does not mention PostgreSQL / CloudBase PG / PG mode /
app.rdb()/queryPgDatabase/managePgDatabase/ RLS
Use this skill when
- You need multi-entity relationship modeling
- You need Mermaid
classDiagramoutput - You want generated model structure and documentation
- You need a clean modeling pass before SQL implementation
How to use this skill (for a coding agent)
-
Clarify the entity set
- Extract business entities, ownership, and relationship cardinality from the request.
- Prefer 3-5 core entities unless the user clearly asks for more.
-
Model first, then generate
- Draft Mermaid
classDiagramcontent. - Validate names, field types, and relationships before calling modeling tools.
- Draft Mermaid
-
Use the right tools
- Read/list existing models ->
manageDataModel(action="list"|"get"|"docs") - Create a new model ->
modifyDataModel(compatibility name; create-only)
- Read/list existing models ->
-
Publish carefully
- Prefer creating with unpublished or draft-like intent first.
- Publish only after checking field names, required constraints, and relationship directions.
Mermaid generation rules
Naming
- Class names -> PascalCase
- Field names -> camelCase
- Convert Chinese business descriptions into clear English identifiers
- Keep enum values human-readable when needed
Type mapping
| Business meaning | Mermaid type |
|---|---|
| text | string |
| number | number |
| boolean | boolean |
| enum | x-enum |
email | |
| phone | phone |
| URL | url |
| image | x-image |
| file | x-file |
| rich text | x-rtf |
| date | date |
| datetime | datetime |
| region | x-area-code |
| location | x-location |
| array | string[] or another explicit array type |
Required structure conventions
- Use
required()only for fields the user explicitly marks as required. - Use
unique()only for explicit uniqueness needs. - Use
display_field()for the human-facing label field. - Add concise
<<description>>notes to important fields. - Keep relationship labels tied to actual field names rather than vague business prose.
Minimal example
classDiagram
class User {
username: string <<Username>>
email: email <<Email>>
display_field() "username"
required() ["username", "email"]
unique() ["username", "email"]
}
class Order {
orderNo: string <<Order Number>>
totalAmount: number <<Total Amount>>
userId: string <<User ID>>
display_field() "orderNo"
unique() ["orderNo"]
}
Order "n" --> "1" User : userId
%% Class naming
note for User "用户"
note for Order "订单"
Tool usage guidance
Read existing models
Use this bef
阅读完整来源 ↗