No cold start
A single long-running process serves every request. Compiled modules are cached, so imports after the first are a map lookup — not a recompile.
Langoost is a statically-typed scripting language that compiles to bytecode and runs on a fast VM. A long-running process serves every request — like PHP, without the per-request warm-up.
Open source and in active development. Build from source with Go today; prebuilt binaries are planned.
// hello.goost
let name: string = "world"
print("Hello, " + name + "!") A single long-running process serves every request. Compiled modules are cached, so imports after the first are a map lookup — not a recompile.
A large standard library ships in the box: strings, math, json, yaml, xml, http, net, crypto, io, exec, collections, and thread.
Python-like simplicity with TypeScript-style annotations. Write let x: int = 1 when it helps, omit it when it doesn't.
Source compiles to a compact 4-byte bytecode that runs on a stack VM. Disassemble any script to see exactly what executes.
Call http.serve with a handler that takes the method,
path, query, and body and returns a string. Every request runs in its
own isolated VM, so concurrent requests never share mutable state.
import http
import json
// Each request runs in its own isolated VM.
fn handle(method: string, path: string, query: string, body: string): string {
if path == "/health" {
return "{\"status\": \"ok\"}"
}
let data: json = {message: "Hello from Langoost"}
return json.pretty(data)
}
http.serve(8080, handle) import http
import json
// Fetch and reshape JSON in a handful of lines.
let body = http.get("https://api.example.com/users/1")
let name = json.get(body, "name")
let out: json = {greeting: "hello", to: name}
print(json.pretty(out))
HTTP, JSON, YAML, XML, crypto, files, sockets, compression, and
concurrency all ship in the box. Inline json literals make
shaping data feel native.
A quick reference while the full specification grows.
Every page ships clean semantic HTML, JSON-LD structured data, and a
plain-text .md mirror. Point an AI assistant at our
llms.txt and it gets the whole language in one fetch.
A few minutes to the shape of Langoost — syntax, types, and the standard library.