Token导航 LogoToken导航TokenDH.com
开发只读github未标认证来源可访问许可证需确认审计通过

database-designer-lite数据库设计器精简版

Agent Skill

用于辅助数据库表结构、查询语句、迁移脚本和数据维护任务。它适合让 Agent 分析 schema、编写 SQL、排查查询问题、整理索引或生成迁移建议。使用时需要明确数据库类型、连接环境和目标表,区分只读分析与写入变更;涉及删除、更新、迁移和批量导入时,应优先 dry-run、备份或事务保护,避免误操作。

总安装

194

周安装

8

GitHub Stars

1

下载量

63
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:database-designer-lite(数据库设计器精简版)
来源仓库:https://github.com/rainfly-code/database-designer-lite
仓库路径:skills/database-designer-lite
安装命令:
npx skills add https://github.com/rainfly-code/database-designer-lite --skill database-designer-lite
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/rainfly-code/database-designer-lite --skill database-designer-lite

简介

面向个人项目的小型化数据库设计助手,兼顾简洁性与规范性。

  • 自动识别核心功能模块并推荐合适的范式化结构与扩展方案。
  • 支持主流关系型数据库(如 MySQL/PostgreSQL),避免过度工程化。
  • 用户需提供项目背景或功能清单作为输入依据。
  • database-designer-lite 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Personal Database Designer

This skill helps users design databases for personal-level projects. It balances simplicity with good design practices, ensuring the database is not over-engineered (avoiding complex enterprise structures) but is robust, normalized, and extensible.

Input

The user will provide:

  1. A requirement description, project name, or design document.
  2. The target database type (e.g., MySQL, PostgreSQL, SQLite). If not specified, ask or default to a common choice like MySQL or PostgreSQL based on context.

Workflow

  1. Analyze System Functions (Step 1):

- Start by understanding the core purpose of the system. - If the user only provides a project name (e.g., "Defect Management System"), infer the key functional modules (e.g., User Management, Project Tracking, Defect Reporting). - Goal: List the high-level functions the system must perform.

  1. Extract Business Entities (Step 2):

- Based on the functions from Step 1, identify the nouns/objects involved. - Example: "User reports a defect in a project" -> Entities: User, Defect, Project. - Goal: Create a list of core business entities.

  1. Derive Entity Relationships (Step 3):

- Determine how the entities interact (the verbs). - Example: One User can report many Defects (1:N); One Project has many Defects (1:N). - Goal: Define the cardinality (1:1, 1:N, N:M) between entities.

  1. Design Database Tables (Step 4):

- Map entities to tables. - Apply Normalization (3NF) rules. - Define columns, primary keys (id), and foreign keys. - Determine data types and constraints. - Goal: Detailed table schema definitions.

  1. Generate SQL (Step 5):

- Write the DDL statements to create the schema. - Include comments for tables and columns. - Goal: Executable SQL script.

  1. Generate Database Design Document (Step 6):

- Compile the results of Steps 1-5 into a structured Markdown document. - Crucial: Use the Write tool to save this document to a file named {project_name}_design.md. - Goal: A final, saved design file containing the overview, relationships, table structures, and SQL.

Design Principles

  1. Complexity & Scale:

- Control Table Count: - Small Projects: 4-6 tables. - Medium Projects: 6-10 tables. - Limit: Avoid exceeding 15 tables for personal-level projects. - Avoid Over-Abstraction: Do NOT create base_entity tables, multi-level inheritance tables, or complex generic relationship tables (EAV model). - One Table = One Entity: Each table must represent a clear business object (e.g., users, projects, tasks, defects, orders). Do NOT mix multiple concepts in one table.

  1. Normalization:

- Follow 3rd Normal Form (3NF): - Each field must describe the current table's entity. - No redundant data storage. - Avoid repeating information. - Denormalize ONLY if there is a clear performance benefit and it's simple to maintain.

  1. Naming Conventions:

- Tables: Plural, snake_case (e.g., users, order_items). - Columns: snake_case (e.g., user_id, created_at). - Primary Keys: id (BIGINT PRIMARY KEY AUTO_INCREMENT). Avoid composite keys or UUIDs unless strictly necessary. - Foreign Keys: related_table_id (e.g., user_id, project_id).

  1. Field Types:

- Status/Enums: ALWAYS use Numeric types (TINYINT or INT) for status, priority, severity, type, etc. NEVER use ENUM or VARCHAR for these fields. - Roles: Use a dedicated roles table and a many-to-many relationship (user_roles table) if roles are dynamic or if users can have multiple roles. Even for simple roles, prefer numeric IDs over strings. - No Field Abuse: Do not use generic columns like data1, data2 or info JSON columns unless the schema is truly dynamic.

  1. Documentation (Comments):

- Crucial: You MUST add SQL comments (COMMENT '...') to every table and every column. - Status Fields: Explicitly explain numeric values (e.g., COMMENT '0: Pending, 1: Active, 2: Closed').

  1. Deletion Strategy:

- Business Data (e.g., Users, Orders, Posts): Use Logical Deletion. Add a column like is_deleted (BOOLEAN, default 0/FALSE) or deleted_at (TIMESTAMP, nullable). - Temporary/Log Data (e.g., Session tokens, Cache, Temporary uploads): Use Physical Deletion. No soft delete columns.

  1. Standard Fields:

- All business tables MUST include created_at and updated_at timestamps.

  1. Relationships:

- One-to-One: Foreign key on the "owner" side, or shared PK. - One-to-Many: Foreign key on the "many" side. - Many-to-Many: ALWAYS use an association/junction table (e.g., user_roles, task_tags).

  1. Indexes:

- Add indexes ONLY for Foreign Keys and frequently queried fields (e.g., email, username, status). Do not over-index.

  1. Extensibility:

- Design should allow adding new features (e.g., new status types, new modules) without major schema refactoring.

Output Format

You must output two distinct sections, and save the entire content to a file.

1. File Output Requirement

  • Action: You MUST use the Write (or equivalent) tool to save the entire content (Design Document + SQL) to a file.
  • Filename: Use {project_name}_design.md (snake_case). If no project name is clear, use database_design.md.

2. Database Design Document (Markdown)

  • Project Overview: Brief summary.
  • Entity Relationships: Explain the relationships between main entities.
  • Table Structures:

- Table Name & Description. - Columns: Name, Type, Constraints, Comments/Description (Explain what numeric values mean). - Deletion Strategy.

3. Executable SQL

  • Provide a single code block with the complete SQL script.
  • Include Comments: Use COMMENT syntax in CREATE TABLE statements for tables and columns.
  • Ensure it runs without errors (e.g., correct ordering for foreign keys).
  • Use the syntax appropriate for the requested database type.

Example Interaction

User: "I need a database for a personal blog. Users can write posts and comment. Use MySQL."

Response:

Database Design: Personal Blog

Tables

  1. users

- Stores user account information. - Strategy: Logical Deletion (deleted_at). - Columns: id, username, email, password_hash, created_at, updated_at, deleted_at.

  1. posts

- Blog entries. - Strategy: Logical Deletion (deleted_at). - Columns: id, user_id (FK), title, content, published_at, created_at, updated_at, deleted_at.

...

SQL Script

-- Users Table
CREATE TABLE users (
    id INT AUTO_INCREMENT PRIMARY KEY,
    username VARCHAR(50) NOT NULL UNIQUE,
    email VARCHAR(100) NOT NULL UNIQUE,
    password_hash VARCHAR(255) NOT NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
    deleted_at TIMESTAMP NULL, -- Logical deletion
    INDEX idx_email (email)
);

-- Posts Table
CREATE TABLE posts (
    id INT AUTO_INCREMENT PRIMARY KEY,
    user_id INT NOT NULL,
    title VARCHAR(255) NOT NULL,
    content TEXT,
    published_at TIMESTAMP NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
    updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
    deleted_at TIMESTAMP NULL, -- Logical deletion
    FOREIGN KEY (user_id) REFERENCES users(id) ON DELETE CASCADE
);

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

38.21%
按下载量换算24

Claude

27.02%
按下载量换算17

Cursor

18.66%
按下载量换算12

Gemini CLI

8.34%
按下载量换算5

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills