name: Platform API Design description: This skill should be used when the user asks to "add endpoint", "create API", "platform backend", "Hono route", "Prisma model", "backend service", mentions "metorial-platform/src/backend/", or discusses TypeScript backend development for the Metorial platform. Provides guidance for building APIs following platform conventions with Hono, Prisma, and Bun. version: 1.0.0

Platform API Design

This skill provides guidance for developing backend APIs in the Metorial platform. The platform uses TypeScript with Hono framework, Prisma ORM, and Bun runtime.

Context Loading

Before implementing platform APIs, gather context from existing tools:

1. Pattern Context (Drift)

Use Drift to understand existing API patterns:

drift_context targeting metorial-platform/src/backend/

This provides detected patterns for:

Route structure and naming
Request/response handling
Authentication middleware
Error handling

2. Workflow Context (Serena)

Use Serena for development workflow:

read_memory('development_workflow')

This provides:

Build commands (mise run platform:build)
Test commands (mise run platform:test)
Prisma commands (mise run platform:prisma:generate)

3. Schema Navigation (Serena)

Navigate existing code with Serena:

find_symbol targeting metorial-platform/src/backend/

Technology Stack

Component	Technology	Purpose
Runtime	Bun	Fast JS/TS runtime
Framework	Hono	Web framework
ORM	Prisma	Database access
Validation	Zod	Schema validation
Auth	JWT	Authentication

Project Structure

metorial-platform/src/backend/
├── routes/              # API route handlers
│   ├── auth/           # Authentication routes
│   ├── users/          # User management
│   └── servers/        # MCP server routes
├── middleware/          # Hono middleware
│   ├── auth.ts         # JWT validation
│   └── error.ts        # Error handling
├── services/            # Business logic
├── lib/                 # Shared utilities
│   ├── prisma.ts       # Prisma client
│   └── validation.ts   # Zod schemas
└── index.ts            # App entry point

Implementation Patterns

Route Definition

Create routes in routes/<domain>/index.ts:

import { Hono } from 'hono';
import { zValidator } from '@hono/zod-validator';
import { z } from 'zod';
import { authMiddleware } from '../../middleware/auth';
import { prisma } from '../../lib/prisma';

const app = new Hono();

// Input validation schema
const CreateItemSchema = z.object({
  name: z.string().min(1).max(100),
  description: z.string().max(1000).optional(),
});

// Protected route with validation
app.post(
  '/',
  authMiddleware,
  zValidator('json', CreateItemSchema),
  async (c) => {
    const { name, description } = c.req.valid('json');
    const userId = c.get('userId');

    const item = await prisma.item.create({
      data: {
        name,
        description,
        userId,
      },
    });

    return c.json(item, 201);
  }
);

export default app;

Prisma Model Definition

Add models to prisma/schema.prisma:

model Item {
  id          String   @id @default(cuid())
  name        String
  description String?
  userId      String
  user        User     @relation(fields: [userId], references: [id])
  createdAt   DateTime @default(now())
  updatedAt   DateTime @updatedAt

  @@index([userId])
}

After modifying schema:

mise run platform:prisma:generate
mise run platform:prisma:push  # or prisma migrate dev

Authentication Middleware

Standard auth pattern:

import { Context, Next } from 'hono';
import { verify } from 'hono/jwt';

export const authMiddleware = async (c: Context, next: Next) => {
  const authHeader = c.req.header('Authorization');

  if (!authHeader?.startsWith('Bearer ')) {
    return c.json({ error: 'Unauthorized' }, 401);
  }

  const token = authHeader.slice(7);

  try {
    const payload = await verify(token, process.env.JWT_SECRET!);
    c.set('userId', payload.sub);
    await next();
  } catch {
    return c.json({ error: 'Invalid token' }, 401);
  }
};

Error Handling

Consistent error responses:

import { Context, Next } from 'hono';
import { HTTPException } from 'hono/http-exception';

export const errorMiddleware = async (c: Context, next: Next) => {
  try {
    await next();
  } catch (error) {
    if (error instanceof HTTPException) {
      return c.json({ error: error.message }, error.status);
    }

    if (error instanceof Prisma.PrismaClientKnownRequestError) {
      if (error.code === 'P2002') {
        return c.json({ error: 'Resource already exists' }, 409);
      }
      if (error.code === 'P2025') {
        return c.json({ error: 'Resource not found' }, 404);
      }
    }

    console.error('Unhandled error:', error);
    return c.json({ error: 'Internal server error' }, 500);
  }
};

API Conventions

URL Structure

/api/v1/<resource>           # Collection
/api/v1/<resource>/:id       # Individual resource
/api/v1/<resource>/:id/sub   # Nested resource

HTTP Methods

Method	Purpose	Response Code
GET	Retrieve resource(s)	200
POST	Create resource	201
PUT	Replace resource	200
PATCH	Update resource	200
DELETE	Remove resource	204

Response Format

Success response:

{
  "data": { ... },
  "meta": {
    "total": 100,
    "page": 1,
    "perPage": 20
  }
}

Error response:

{
  "error": "Error message",
  "code": "ERROR_CODE",
  "details": { ... }
}

Pagination

Use cursor-based pagination for lists:

const PaginationSchema = z.object({
  cursor: z.string().optional(),
  limit: z.coerce.number().min(1).max(100).default(20),
});

app.get('/', zValidator('query', PaginationSchema), async (c) => {
  const { cursor, limit } = c.req.valid('query');

  const items = await prisma.item.findMany({
    take: limit + 1,
    cursor: cursor ? { id: cursor } : undefined,
    orderBy: { createdAt: 'desc' },
  });

  const hasMore = items.length > limit;
  const data = hasMore ? items.slice(0, -1) : items;

  return c.json({
    data,
    meta: {
      nextCursor: hasMore ? data[data.length - 1].id : null,
      hasMore,
    },
  });
});

Validation Patterns

Request Validation

Use Zod for all request validation:

// Define schemas in lib/validation/
export const schemas = {
  // Path params
  IdParam: z.object({
    id: z.string().cuid(),
  }),

  // Query params
  SearchQuery: z.object({
    q: z.string().min(1).max(100),
    page: z.coerce.number().positive().default(1),
  }),

  // Body schemas
  CreateUser: z.object({
    email: z.string().email(),
    name: z.string().min(1).max(100),
    password: z.string().min(8),
  }),
};

Response Validation

For type safety, define response types:

// Types derived from Prisma
type User = Prisma.UserGetPayload<{
  select: {
    id: true;
    email: true;
    name: true;
    createdAt: true;
  };
}>;

// API response type
interface ApiResponse<T> {
  data: T;
  meta?: {
    total?: number;
    page?: number;
    perPage?: number;
  };
}

Service Layer Pattern

Separate business logic from routes:

// services/user.service.ts
import { prisma } from '../lib/prisma';
import { hash, verify } from '../lib/auth';

export const userService = {
  async create(data: CreateUserInput) {
    const hashedPassword = await hash(data.password);

    return prisma.user.create({
      data: {
        ...data,
        password: hashedPassword,
      },
      select: {
        id: true,
        email: true,
        name: true,
      },
    });
  },

  async findById(id: string) {
    return prisma.user.findUnique({
      where: { id },
      select: {
        id: true,
        email: true,
        name: true,
      },
    });
  },

  async authenticate(email: string, password: string) {
    const user = await prisma.user.findUnique({ where: { email } });

    if (!user || !(await verify(password, user.password))) {
      return null;
    }

    return { id: user.id, email: user.email };
  },
};

Testing Patterns

Route Testing

import { describe, it, expect, beforeAll } from 'bun:test';
import app from './index';

describe('POST /api/v1/items', () => {
  let authToken: string;

  beforeAll(async () => {
    // Setup test auth token
    authToken = await getTestToken();
  });

  it('creates item with valid input', async () => {
    const res = await app.request('/api/v1/items', {
      method: 'POST',
      headers: {
        'Authorization': `Bearer ${authToken}`,
        'Content-Type': 'application/json',
      },
      body: JSON.stringify({
        name: 'Test Item',
        description: 'Description',
      }),
    });

    expect(res.status).toBe(201);
    const data = await res.json();
    expect(data.name).toBe('Test Item');
  });

  it('rejects invalid input', async () => {
    const res = await app.request('/api/v1/items', {
      method: 'POST',
      headers: {
        'Authorization': `Bearer ${authToken}`,
        'Content-Type': 'application/json',
      },
      body: JSON.stringify({
        name: '', // Invalid: empty
      }),
    });

    expect(res.status).toBe(400);
  });
});

Validation Checklist

Before submitting API changes:

Routes follow URL conventions
All inputs validated with Zod
Authentication applied where needed
Error handling covers edge cases
Prisma migrations created if schema changed
Tests cover success and error cases
Pattern compliance verified (drift_validate_change)

Additional Resources

Reference Files

For detailed patterns and examples:

references/backend-conventions.md - Deep backend patterns

External References

Serena memory: development_workflow for build/test commands
Drift patterns: drift_context for existing conventions
Platform source: metorial-platform/src/backend/

platform-api-design

SKILL.md