bobmatnyc/claude-mpm-skills2 files

Hono Rpc

Hono RPC - end-to-end type-safe API client generation with hc client and TypeScript inference

Specification
Skill ID
bobmatnyc/claude-mpm-skills/hono-rpc
Publisher
bobmatnyc
Repository
claude-mpm-skills
Installs
0
Files
2
Synced
Sep 23, 2026
How to use it

Open any RiverX project, open the Skills panel in the chat, and search for this identifier. The files are fetched from the source repository at install time.

bobmatnyc/claude-mpm-skills/hono-rpcInstalls these files
  • SKILL.md
  • metadata.json

What this skill tells the agent

Hono RPC - Type-Safe Client

Overview

Hono RPC enables sharing API specifications between server and client through TypeScript's type system. Export your server's type, and the client automatically knows all routes, request shapes, and response types - no code generation required.

Key Features:

  • Zero-codegen type-safe client
  • Automatic TypeScript inference
  • Works with Zod validators
  • Status code-aware response types
  • Supports path params, query, headers

When to Use This Skill

Use Hono RPC when:

  • Building full-stack TypeScript applications
  • Need type-safe API consumption without OpenAPI/codegen
  • Want compile-time validation of API calls
  • Sharing types between client and server in monorepos

Basic Setup

Server Side

// server/index.ts
import { Hono } from 'hono'
import { zValidator } from '@hono/zod-validator'
import { z } from 'zod'

const app = new Hono()

// Define routes with validation
const route = app
  .get('/users', async (c) => {
    const users = [{ id: '1', name: 'Alice' }]
    return c.json({ users })
  })
  .post(
    '/users',
    zValidator('json', z.object({
      name: z.string(),
      email: z.string().email()
    })),
    async (c) => {
      const data = c.req.valid('json')
      return c.json({ id: '1', ...data }, 201)
    }
  )
  .get('/users/:id', async (c) => {
    const id = c.req.param('id')
    return c.json({ id, name: 'Alice' })
  })

// Export type for client
export type AppType = typeof route

export default app

Client Side

// client/api.ts
import { hc } from 'hono/client'
import type { AppType } from '../server'

// Create typed client
const client = hc<AppType>('http://localhost:3000')

// All methods are type-safe!
async function examples() {
  // GET /users
  const usersRes = await client.users.$get()
  const { users } = await usersRes.json()
  // users: { id: string; name: string }[]

  // POST /users - body is typed
  const createRes = await client.users.$post({
    json: {
      name: 'Bob',
      email: 'bob@example.com'
    }
  })
  const created = await createRes.json()
  // created: { id: string; name: string; email: string }

  // GET /users/:id - params are typed
  const userRes = await client.users[':id'].$get({
    param: { id: '123' }
  })
  const user = await userRes.json()
  // user: { id: string; name: string }
}

Route Chaining for Type Export

Important: Chain routes for proper type inference:

// CORRECT: Chain all routes
const route = app
  .get('/a', handlerA)
  .post('/b', handlerB)
  .get('/c', handlerC)

export type AppType = typeof route

// WRONG: Separate statements lose type info
app.get('/a', handlerA)
app.post('/b', handlerB)  // Types lost!

export type AppType = typeof app  // Missing routes!

Request Patterns

Path Parameters

// Server
const route = app.get('/posts/:postId/comments/:commentId', async (c) => {
  const { postId, commentId } = c.req.param()
  return c.json({ postId, commentId })
})

// Client
const res = await client.posts[':postId'].comments[':commentId'].$get({
  param: {
    postId: '1',
    commentId: '42'
  }
})

Query Parameters

// Server
const route = app.get(
  '/search',
  zValidator('query', z.object({
    q: z.string(),
    page: z.coerce.number().optional(),
    limit: z.coerce.number().optional()
  })),
  async (c) => {
    const { q, page, limit } = c.req.valid('query')
    return c.json({ query: q, page, limit })
  }
)

// Client
const res = await client.search.$get({
  query: {
    q: 'typescript',
    page: 1,
    limit: 20
  }
})

JSON Body

// Server
const route = app.post(
  '/posts',
  zValidator('json', z.object({
    title: z.string(),
    content: z.string(),
    tags: z.array(z.string()).optional()
  })),
  async (c) => {
    const data = c.req.valid('json')
    return c.json({ id: '1', ...data }, 201)
  }
)

// Client
const res = await client.posts.$post({
  json: {
    title: 'Hello World',
    content: 'My first post',
    tags: ['typescript', 'hono']
  }
})

Form Data

// Server
const route = app.post(
  '/upload',
  zValidator('form', z.object({
    file: z.instanceof(File),
    description: z.string().optional()
  })),
  async (c) => {
    const { file, description } = c.req.valid('form')
    return c.json({ filename: file.name })
  }
)

// Client
const formData = new FormData()
formData.append('file', file)
formData.append('description', 'My file')

const res = await client.upload.$post({
  form: formData
})

Headers

// Server
const route = app.get(
  '/protected',
  zValidator('header', z.object({
    authorization: z.string()
  })),
  async (c) => {
    return c.json({ authenticated: true })
  }
)

// Client
const res = await client.protected.$get({
  header: {
    authorization: 'Bearer token123'
  }
})

Response Type Inference

Status Code-Aware Types

// Server
const route = app.get('/user', async (c) => {
  const user = await getUser()

  if (!user) {
    return c.json({ error: 'Not found' }, 404)
  }

  return c.json({ id: user.id, name: user.name }, 200)
})

// Client - use InferResponseType
import { InferResponseType } from 'hono/client'

type SuccessResponse = InferResponseType<typeof client.user.$get, 200>
// { id: string; name: string }

type ErrorResponse = InferResponseType<typeof client.user.$get, 404>
// { error: string }

// Handle different status codes
const res = await client.user.$get()

if (res.status === 200) {
  const data = await res.json()
  // data: { id: string; name: string }
} else if (res.status === 404) {
  const error = await res.json()
  // error: { error: string }
}

Request Type Inference

import { InferRequestType } from 'hono/client'

type CreateUserRequest = InferRequestType<typeof client.users.$post>['json']
// { name: string; email: string }

// Use for form validation, state management, etc.
const [formData, setFormData] = useState<CreateUserRequest>({
  name: '',
  email: ''
})

Multi-File Route Organization

Organize Routes

// server/routes/users.ts
import { Hono } from 'hono'

export const users = new Hono()
  .get('/', async (c) => c.json({ users: [] }))
  .post('/', async (c) => c.json({ created: true }, 201))
  .get('/:id', async (c) => c.json({ id: c.req.param('id') }))

// server/routes/posts.ts
export const posts = new Hono()
  .get('/', async (c) => c.json({ posts: [] }))
  .post('/', async (c) => c.json({ created: true }, 201))

// server/index.ts
import { Hono } from 'hono'
import { users } from './routes/users'
import { posts } from './routes/posts'

const app = new Hono()

const route = app
  .route('/users', users)
  .route('/posts', posts)

export type AppType = typeof route
export default app

Client Usage

import { hc } from 'hono/client'
import type { AppType } from '../server'

const client = hc<AppType>('http://localhost:3000')

// Routes are nested
await client.users.$get()         // GET /users
await client.users[':id'].$get()  // GET /users/:id
await client.posts.$get()         // GET /posts

Error Handling

Handle Fetch Errors

async function fetchUser(id: string) {
  try {
    const res = await client.users[':id'].$get({
      param: { id }
    })

    if (!res.ok) {
      const error = await res.json()
      throw new Error(error.message || 'Failed to fetch user')
    }

    return await res.json()
  } catch (error) {
    if (error instanceof TypeError) {
      // Network error
      throw new Error('Network error')
    }
    throw error
  }
}

Type-Safe Error Responses

// Server
const route = app.get('/resource', async (c) => {
  try {
    const data = await fetchData()
    return c.json({ success: true, data })
  } catch (e) {
    return c.json({ success: false, error: 'Failed' }, 500)
  }
})

// Client

Use this skill in a project

Open any RiverX project, search the Skills panel for bobmatnyc/claude-mpm-skills/hono-rpc, and install it in one click.

Use this skill