npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@openuidev/svelte-lang

v0.1.0

Published

Define component libraries, generate LLM system prompts, and render streaming OpenUI Lang output in Svelte 5 — the Svelte runtime for OpenUI generative UI

Readme

@openuidev/svelte-lang

Svelte 5 runtime for OpenUI — define component libraries, generate model prompts, and render structured UI from streaming LLM output.

License: MIT

Install

npm install @openuidev/svelte-lang
# or
pnpm add @openuidev/svelte-lang

Peer dependencies: svelte >=5.0.0

Overview

@openuidev/svelte-lang provides three core capabilities:

  1. Define components — Use defineComponent and createLibrary to declare what the model is allowed to generate, with typed props via Zod schemas.
  2. Generate prompts — Call library.prompt() to produce a system prompt that instructs the model how to emit OpenUI Lang output.
  3. Render output — Use <Renderer> to parse and progressively render streamed OpenUI Lang into Svelte components.

Quick Start

1. Define a component

<!-- Greeting.svelte -->
<script lang="ts">
  import type { ComponentRenderProps } from "@openuidev/svelte-lang";

  let { props, renderNode }: ComponentRenderProps<{ name: string; mood?: "happy" | "excited" }> = $props();
</script>

<div class={props.mood === "excited" ? "text-xl font-bold" : ""}>
  Hello, {props.name}!
</div>
import { defineComponent } from "@openuidev/svelte-lang";
import { z } from "zod";
import Greeting from "./Greeting.svelte";

const GreetingDef = defineComponent({
  name: "Greeting",
  description: "Displays a greeting message",
  props: z.object({
    name: z.string().describe("The person's name"),
    mood: z.enum(["happy", "excited"]).optional().describe("Tone of the greeting"),
  }),
  component: Greeting,
});

2. Create a library

import { createLibrary } from "@openuidev/svelte-lang";

const library = createLibrary({
  components: [GreetingDef, CardDef, TableDef /* ... */],
  root: "Card", // optional default root component
});

3. Generate a system prompt

const systemPrompt = library.prompt({
  preamble: "You are a helpful assistant.",
  additionalRules: ["Always greet the user by name."],
  examples: ['User: Hi\n\nroot = Greeting("Alice", "happy")'],
});

4. Render streamed output

<script lang="ts">
  import { Renderer } from "@openuidev/svelte-lang";
  import { library } from "$lib/library";

  let { response, isStreaming }: { response: string | null; isStreaming: boolean } = $props();
</script>

<Renderer
  {response}
  {library}
  {isStreaming}
  onAction={(event) => console.log("Action:", event)}
/>

API Reference

Component Definition

| Export | Description | | :-------------------------- | :----------------------------------------------------------------------------------------- | | defineComponent(config) | Define a single component with a name, Zod props schema, description, and Svelte component | | createLibrary(definition) | Create a library from an array of defined components |

Rendering

| Export | Description | | :--------- | :---------------------------------------------------------- | | Renderer | Svelte component that parses and renders OpenUI Lang output |

RendererProps:

| Prop | Type | Description | | :-------------- | :-------------------------------------- | :---------------------------------------------------------------- | | response | string \| null | Raw OpenUI Lang text from the model | | library | Library | Component library from createLibrary() | | isStreaming | boolean | Whether the model is still streaming (disables form interactions) | | onAction | (event: ActionEvent) => void | Callback when a component triggers an action | | onStateUpdate | (state: Record<string, any>) => void | Callback when form field values change | | initialState | Record<string, any> | Initial form state for hydration | | onParseResult | (result: ParseResult \| null) => void | Callback when the parse result changes |

Children Rendering

Svelte components receive renderNode as a snippet prop (not via context). Use it to render child element nodes:

<script lang="ts">
  import type { Snippet } from "svelte";

  let { props, renderNode }: { props: { children?: unknown }; renderNode: Snippet<[unknown]> } = $props();
</script>

<div>
  {#if props.children}
    {@render renderNode(props.children)}
  {/if}
</div>

Parser (Server-Side)

| Export | Description | | :------------------------------ | :----------------------------------------------------- | | createParser(schema) | Create a one-shot parser for complete OpenUI Lang text | | createStreamingParser(schema) | Create an incremental parser for streaming input |

The streaming parser exposes two methods:

| Method | Description | | :------------ | :---------------------------------------------------- | | push(chunk) | Feed the next chunk; returns the latest ParseResult | | getResult() | Get the latest result without consuming new data |

After the stream ends, check meta.unresolved for any identifiers that were referenced but never defined. During streaming these are expected (forward refs) and are not treated as errors.

Errors

ParseResult.meta.errors contains structured OpenUIError objects. Each error has a type discriminant (currently always "validation") and a code for consumer-side filtering:

| Code | Meaning | | :------------------ | :-------------------------------------------------- | | missing-required | Required prop absent with no default | | null-required | Required prop explicitly null with no default | | unknown-component | Component name not found in the library schema | | excess-args | More positional args passed than the schema defines |

Errors do not affect rendering — the parser stays permissive and renders what it can:

const result = parser.parse(output);
const critical = result.meta.errors.filter((e) => e.code === "unknown-component");

Context Getters

Use these inside component renderers to interact with the rendering context:

| Function | Returns | Description | | :------------------------- | :-------------------- | :----------------------------------------------------------------------------- | | getOpenUIContext() | OpenUIContextValue | Access the full context object (library, streaming state, field accessors) | | getIsStreaming() | () => boolean | Returns a getter for the streaming state — call it reactively: isStreaming() | | getTriggerAction() | Function | Trigger an action event | | getGetFieldValue() | Function | Get a form field's current value | | getSetFieldValue() | Function | Set a form field's value | | getFormName() | string \| undefined | Get the current form's name | | useSetDefaultValue(opts) | void | Set a field's default value once streaming completes |

Form Validation

| Export | Description | | :--------------------------- | :---------------------------------------------------- | | getFormValidation() | Access form validation state | | createFormValidation() | Create a form validation context | | setFormValidationContext() | Provide validation context to child components | | validate(value, rules) | Run validation rules against a value | | builtInValidators | Built-in validators (required, email, min, max, etc.) | | parseRules(rules) | Parse a rules config object into ParsedRule[] |

Types

import type {
  // Component definition
  Library,
  LibraryDefinition,
  DefinedComponent,
  ComponentRenderer,
  ComponentRenderProps,
  ComponentGroup,
  SubComponentOf,
  PromptOptions,

  // Rendering
  RendererProps,
  OpenUIContextValue,
  ActionConfig,

  // Parser & core
  ActionEvent,
  ElementNode,
  ParseResult,
  LibraryJSONSchema,

  // Validation
  FormValidationContextValue,
  ParsedRule,
  ValidatorFn,
} from "@openuidev/svelte-lang";

JSON Schema Output

Libraries can also produce a JSON Schema representation of their components:

const schema = library.toJSONSchema();
// schema.$defs["Card"]     → { properties: {...}, required: [...] }
// schema.$defs["Greeting"] → { properties: {...}, required: [...] }

Differences from React

| Concern | react-lang | svelte-lang | | :----------------- | :----------------------------------------------- | :----------------------------------------------- | | Children rendering | renderNode function prop returning ReactNode | renderNode snippet (Snippet<[unknown]>) | | Context access | Hooks (useIsStreaming(), etc.) | Getters (getIsStreaming(), etc.) | | Error boundaries | Class-based, preserves last valid render | svelte:boundary with auto-retry on prop change | | Reactivity | Hooks re-run on every render | Runes ($state, $derived, $effect) |

Documentation

Full documentation, guides, and the language specification are available at openui.com.

License

MIT