How to use the Backend Patterns (Japanese) agent skill

What the skill makes the agent do

Agent skills are instruction files that an AI agent loads when a task matches them. This skill's SKILL.md covers backend architecture patterns, API design, database optimization and server-side best practices. The examples are in TypeScript and use Supabase, Redis and Next.js.

When the skill applies, the agent tends to follow these conventions: RESTful URL design (for example GET /api/markets and GET /api/markets/:id), a repository layer that hides data access behind an interface, a service layer that holds business logic, auth middleware that reads a Bearer token from the Authorization header, selecting only the database columns you need, batch fetching to avoid N+1 queries, Redis caching with a 5-minute expiry, and centralized error handling that returns a consistent JSON shape.

Install the backend-patterns skill

Download the skill ZIP from this page and unzip it. The skill folder contains SKILL.md. Put the folder in one of these locations:

You can install it for your user account in ~/.claude/skills/, or for a single project in that project's .claude/skills/ folder. Alternatively, in the Claude apps go to Settings → Capabilities → Skills and upload the ZIP.

• Personal: ~/.claude/skills/
• Project: .claude/skills/ inside your project folder
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP

Trigger the skill with natural requests

You do not need a special command. Describe the backend task in plain language and mention the kind of work involved, such as API design, caching, database queries or error handling. The agent decides whether the skill fits your request.

Be specific about your stack. The patterns are written for Node.js, Express and Next.js API routes, so naming those helps the agent apply them correctly.

• Design a REST API for a list of products with filtering, sorting and pagination in Next.js.
• Add Redis caching with a 5-minute expiry to my user lookup function.
• Review this API route for N+1 queries and suggest a batch fetch.
• Add a withAuth wrapper that checks a Bearer token on my Next.js API route.

What to expect from the output

The skill gives the agent patterns to follow, not a finished production system. The agent will typically produce code shaped like the examples in the guide: a repository interface and implementation, a service class, a cache-aside function, or a centralized error handler.

Check the output against your own project. Table names, types and helper functions such as verifyToken or generateEmbedding are placeholders in the source and must be implemented for your code.

Safety notes before you rely on it

According to the skill's description, it runs no scripts and uses no network access. It is a set of instructions, so you still review any code the agent writes before running it.

The authentication and error examples are illustrative. The guide does not cover token validation details, rate limiting or input sanitisation in depth. Do not treat the sample JWT check as a complete security decision. Store secrets such as the JWT secret in environment variables, as the example does with process.env.JWT_SECRET, and review any database function, such as the Supabase SQL example, before deploying it.

• The skill runs no scripts.
• The skill uses no network access.
• Auth and error examples are illustrative, not a complete security review.
• Rate limiting and input sanitisation are not covered in depth.