Kkrpc

What it does

When your agent uses it

• Building TypeScript RPC with kkrpc stable APIs
• Choosing native Transport<RPCMessage adapters
• Integrating validation
• Remote references

Example prompts

• “/kkrpc”

Requirements

• Node.js
• Compatibility (from SKILL.md): Works in Node.js, Deno, Bun, browsers, Electron, Tauri, Chrome extensions, and queue-backed runtimes with the correct stable entry point and transport.

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Pick the stable entry point.
2. 2Pick the native transport factory.
3. 3Decide whether one side or both sides expose APIs.
4. 4Add plugins only when validation, middleware, logging, or custom behavior is required.

What it can do on your machine

Read from SKILL.md and the folder at commit e558873. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are typescript).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Works in Node.js, Deno, Bun, browsers, Electron, Tauri, Chrome extensions, and queue-backed runtimes with the correct stable entry point and transport.

  From compatibility in the SKILL.md frontmatter.

Context cost

Kkrpc loads about 3.1k tokens when it runs. Until then it costs about 56 tokens; SKILL.md has 618 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~56

When it runs · the whole SKILL.md, loaded when a task matches

~3.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from kunkunsh/kkrpc at commit e558873, republished under its MIT licence (© kunkunsh). 618 words, ~3,058 tokens.

name

kkrpc

description

Use when building TypeScript RPC with kkrpc stable APIs, choosing native Transport<RPCMessage> adapters, or integrating validation, middleware, transferables, streaming, remote references, relay, or inspector tooling.

compatibility

Works in Node.js, Deno, Bun, browsers, Electron, Tauri, Chrome extensions, and queue-backed runtimes with the correct stable entry point and transport.

version

2.0.0

license

MIT

metadata.author

kkrpc

metadata.domain

typescript-rpc

metadata.tags

rpc, typescript, bidirectional, ipc, websocket, http, workers, electron, tauri, chrome-extension

import { expose, wrap } from "kkrpc"

const controller = expose(localAPI, serverTransport)
const remote = wrap<RemoteAPI>(clientTransport)

controller.dispose()

import { RPCChannel } from "kkrpc"

const channel = new RPCChannel<LocalAPI, RemoteAPI>(transport, { expose: localAPI })
const remote = channel.getAPI()

type ServerAPI = {
	math: {
		add(a: number, b: number): Promise<number>
	}
	version(): Promise<string>
}

type ClientAPI = {
	notify(message: string): Promise<void>
}

const clientAPI: ClientAPI = {
	async notify(message) {
		console.log(message)
	}
}

const channel = new RPCChannel<ClientAPI, ServerAPI>(transport, { expose: clientAPI })
const server = channel.getAPI()

console.log(await server.math.add(1, 2))

const counter = await api.counter
const nested = await api.nested.deepObj.prop

await api.process("input", (progress) => {
	console.log("progress", progress)
})

import { expose, wrap } from "kkrpc/streaming"

type LogAPI = {
	tail(service: string): AsyncIterable<string>
}

for await (const line of wrap<LogAPI>(transport).tail("api")) {
	console.log(line)
}

import { proxy, releaseProxy, wrap } from "kkrpc/remote-refs"

const result = await wrap<RemoteAPI>(transport).useCallback(
	proxy(async (value) => `callback:${value}`)
)

const counter = await wrap<RemoteAPI>(transport).createCounter()
await releaseProxy(counter)

import { wrap } from "kkrpc"
import { workerTransport } from "kkrpc/worker"

const worker = new Worker(new URL("./worker.ts", import.meta.url), { type: "module" })
const api = wrap<WorkerAPI>(workerTransport(worker))

console.log(await api.ping())

import { expose } from "kkrpc"
import { workerSelfTransport } from "kkrpc/worker"

expose(
	{
		async ping() {
			return "pong"
		}
	},
	workerSelfTransport()
)

import { wrap } from "kkrpc"
import { webSocketClientTransport } from "kkrpc/ws"

const api = wrap<RemoteAPI>(webSocketClientTransport("ws://localhost:3000/rpc"))

import { expose, wrap } from "kkrpc"
import { validationPlugin } from "kkrpc/validation"

const plugins = [validationPlugin(validators)]
const controller = expose(localAPI, serverTransport, { plugins })
const remote = wrap<RemoteAPI>(clientTransport, { plugins })

import { middlewarePlugin } from "kkrpc/middleware"

const plugins = [
	middlewarePlugin([
		async (ctx, next) => {
			console.log(ctx.method)
			return next()
		}
	])
]

import { superJsonCodec } from "kkrpc/superjson"
import { createTransport } from "kkrpc/transport"

const transport = createTransport({ platform, codec: superJsonCodec() })

import { createInspector, MemoryBackend } from "kkrpc/inspector"

const memory = new MemoryBackend()
const inspector = createInspector({ backends: [memory], options: { trackLatency: true } })
const channel = new RPCChannel<LocalAPI, RemoteAPI>(transport, {
	expose: localAPI,
	plugins: [inspector.plugin("server")]
})

import { relayTransport } from "kkrpc/relay"

const relay = relayTransport(leftTransport, rightTransport)
relay.dispose()

© kunkunsh, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/kkrpc of kunkunsh/kkrpc.

Open the folder on GitHubat commit e558873