菜单
Design production-ready database schemas for SQL and NoSQL databases. Covers normalization, indexing strategy, migration management with rollback safety, query optimization, and multi-tenant patterns. Supports PostgreSQL, MySQL, SQLite, MongoDB, and Vitess.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/JPeetz/agent-skills --skill database-schema-designer命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Design-first API development skill. Generates OpenAPI 3.1 specifications, enforces REST design best practices, validates endpoints, handles versioning, pagination, error formatting, authentication patterns, rate limiting, and idempotency. Activates when users say "design an API", "create OpenAPI spec", "API endpoint", "REST API design", "API contract", "Swagger/OpenAPI doc", "API versioning", "rate limiting", or "API security". Covers REST, GraphQL schema design, and gRPC proto generation with cross-protocol consistency.
AI-powered GraphQL API design, implementation, and optimization. Covers schema-first design, resolver architecture, query optimization with DataLoader for N+1 prevention, mutation patterns with idempotency, real-time subscriptions, Apollo Federation for distributed graphs, security hardening (depth limiting, rate limiting, authz), and production performance (persisted queries, caching, CDN integration). Primary keyword clusters: GraphQL schema design best practices, Apollo Federation subgraph patterns, DataLoader N+1 query optimization, GraphQL security depth limiting rate limiting, GraphQL persisted queries performance, GraphQL subscription real-time patterns, GraphQL error handling union types, GraphQL pagination relay cursor connection, GraphQL caching strategies production, GraphQL resolver architecture patterns. Designed for agentic platforms — Claude Code, Codex, Cursor, Gemini CLI, OpenClaw, GitHub Copilot, Windsurf, and OpenCode.
Use this skill when an agent needs to call a website's hidden or undocumented API: capture real browser requests into a HAR file, derive the exact request shape, build a clean replayable client, verify it outside the browser, and reuse the verified client. Activates when a backend has no documented API, or when you must reproduce the precise request (method, URL, headers, JSON body) of an action you can only perform in a browser. Authorized use only — never to bypass auth or bot detection.
正在显示 SKILL.md
| name | database-schema-designer |
| description | Design production-ready database schemas for SQL and NoSQL databases. Covers normalization, indexing strategy, migration management with rollback safety, query optimization, and multi-tenant patterns. Supports PostgreSQL, MySQL, SQLite, MongoDB, and Vitess. |
| version | 1.0.0 |
| platforms | ["openclaw","claude","codex","cursor","gemini","copilot","opencode","windsurf"] |
| author | {"name":"Skill Foundry (Forge)","source":"PlanetScale Database Skills, Supabase Agent Skills, softaworks/agent-toolkit"} |
| license | MIT |
| risk_tier | L2 |
| tags | ["database","schema","sql","nosql","postgres","mysql","migration","optimization","design"] |
| requires | {"binaries":[]} |
Design production-ready database schemas with built-in best practices. Covering SQL normalization, indexing strategy, migration management with safe rollback patterns, query optimization, and multi-tenant architecture patterns across PostgreSQL, MySQL, SQLite, MongoDB, and Vitess.
Use this skill when:
Database operations can be destructive. This skill enforces:
Start with entities, not tables. Map the domain before writing DDL:
DOMAIN CANVAS:
├── Entities: What things exist? (User, Order, Product, Invoice)
├── Relationships: How do they connect? (one-to-many, many-to-many)
├── Attributes: What properties do they have?
├── Constraints: What must always be true?
├── Access Patterns: What queries will run most often?
└── Growth Projections: How many rows? At what rate?
Normalization checklist:
Example: E-Commerce Schema
-- Users table (normalized, with soft delete)
CREATE TABLE users (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
email TEXT NOT NULL UNIQUE,
name TEXT NOT NULL,
created_at TIMESTAMPTZ NOT NULL DEFAULT now(),
updated_at TIMESTAMPTZ NOT NULL DEFAULT now(),
deleted_at TIMESTAMPTZ -- soft delete
);
CREATE INDEX idx_users_email ON users(email) WHERE deleted_at IS NULL;
-- Products with inventory tracking
CREATE TABLE products (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
sku TEXT NOT NULL UNIQUE,
name TEXT NOT NULL,
price_cents INTEGER NOT NULL CHECK (price_cents >= 0),
inventory_count INTEGER NOT NULL DEFAULT 0,
created_at TIMESTAMPTZ NOT NULL DEFAULT now()
);
CREATE INDEX idx_products_sku ON products(sku);
-- Orders with status state machine
CREATE TYPE order_status AS ENUM (
'pending', 'confirmed', 'processing', 'shipped', 'delivered', 'cancelled'
);
CREATE TABLE orders (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID NOT NULL REFERENCES users(id),
status order_status NOT NULL DEFAULT 'pending',
total_cents INTEGER NOT NULL DEFAULT 0,
shipping_address JSONB NOT NULL,
created_at TIMESTAMPTZ NOT NULL DEFAULT now()
);
CREATE INDEX idx_orders_user_id ON orders(user_id);
CREATE INDEX idx_orders_status ON orders(status)
WHERE status IN ('pending', 'processing');
-- Order items (many-to-many with quantities)
CREATE TABLE order_items (
order_id UUID NOT NULL REFERENCES orders(id) ON DELETE CASCADE,
product_id UUID NOT NULL REFERENCES products(id),
quantity INTEGER NOT NULL CHECK (quantity > 0),
unit_price_cents INTEGER NOT NULL,
PRIMARY KEY (order_id, product_id)
);
The Index Selection Algorithm:
Index types by database:
| Database | Index Types | When to Use |
|---|---|---|
| PostgreSQL | B-tree (default), Hash, GiST, GIN, BRIN, SP-GiST | B-tree for equality/range; GIN for full-text/arrays; BRIN for large sequential data |
| MySQL/InnoDB | B-tree (clustered PK), Full-text, Spatial | B-tree for most cases; Full-text for text search |
| SQLite | B-tree (default) | All standard cases |
| MongoDB | Single-field, Compound, Multikey, Text, Geospatial, Hashed, TTL | Compound for common queries; TTL for expiring data |
Index anti-patterns:
pg_stat_user_indexes or sys.dm_db_index_usage_statsMigration file structure:
-- migrations/001_add_user_preferences.sql
-- UP: Forward migration (what to apply)
-- DOWN: Rollback migration (how to undo)
-- UP MIGRATION
BEGIN;
CREATE TABLE user_preferences (
user_id UUID PRIMARY KEY REFERENCES users(id) ON DELETE CASCADE,
theme TEXT NOT NULL DEFAULT 'light',
notifications JSONB NOT NULL DEFAULT '{}',
updated_at TIMESTAMPTZ NOT NULL DEFAULT now()
);
-- Backfill existing users with defaults
INSERT INTO user_preferences (user_id)
SELECT id FROM users
ON CONFLICT (user_id) DO NOTHING;
COMMIT;
-- DOWN MIGRATION
-- BEGIN;
-- DROP TABLE IF EXISTS user_preferences;
-- COMMIT;
Safe migration rules:
The EXPLAIN-based optimization workflow:
EXPLAIN (ANALYZE, BUFFERS) on the slow queryANALYZE table_nameCommon optimization patterns:
| Problem | Symptom | Fix |
|---|---|---|
| Missing index | Seq Scan on large table | Add covering index for WHERE + JOIN + SELECT |
| N+1 queries | Many small queries | Use JOIN or batch load |
| Over-fetching | SELECT * on wide tables | Select only needed columns |
| Lock contention | UPDATE waiting on locks | Batch updates, use SKIP LOCKED |
| Statistics stale | Planner choosing wrong plan | ANALYZE table; adjust default_statistics_target |
| Pattern | Description | Best For | Trade-offs |
|---|---|---|---|
| Database per tenant | Separate DB per customer | High security, compliance | Many connections, harder cross-tenant queries |
| Schema per tenant | Separate schema per customer | Medium security, shared DB | Easier backups, moderate isolation |
| Row-level (shared) | tenant_id column everywhere | Simplicity, shared resources | Weakest isolation, query complexity |
| Hybrid | Combine patterns by tier | Enterprise customers get isolation | Operational complexity |
// Denormalized by access pattern
db.products.insertOne({
_id: ObjectId(),
name: "Widget",
price: 9.99,
// Embedded reviews (accessed together)
reviews: [
{ user: "alice", rating: 5, comment: "Great!" },
{ user: "bob", rating: 4, comment: "Good value" }
],
// Reference pattern for frequently-updated data
inventory_warehouse_id: ObjectId("...")
});
MongoDB data modeling rules:
# Session store with TTL
SETEX session:abc123 3600 '{"user_id": 42, "role": "admin"}'
# Rate limiting with sliding window
INCR rate:user:42
EXPIRE rate:user:42 60
# Leaderboard with sorted sets
ZADD leaderboard:weekly 1000 player:42 950 player:17
ZREVRANGE leaderboard:weekly 0 9 WITHSCORES