molykit

Build cross-platform AI chat UIs with Makepad and SSE streaming.

Updated Feb 3, 2026
One-click install
npx skills add https://github.com/RoooyHe/betula --skill molykit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: molykit
Source: https://github.com/RoooyHe/betula/tree/main/.claude/skills/molykit
Command: npx skills add https://github.com/RoooyHe/betula --skill molykit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Molykit provides a reusable toolkit for building AI chat interfaces with Makepad, removing boilerplate and enabling cross-platform async UI flows for native and WASM.

Core Features & Use Cases

  • Cross-platform async utilities (PlatformSend, spawn, ThreadToken) to run AI chat tasks both natively and in WASM.
  • Ready-to-use chat widgets (Chat, Messages, PromptInput, Avatar) to compose AI chat UIs quickly.
  • BotClient trait for integrating AI providers and a sharable BotContext for sync UI access.
  • OpenAI-compatible client with SSE streaming support for streaming AI responses.
  • Protocol types for messages, bots, tool calls, and MCP integration for extensible context.
  • Use cases include building chat interfaces, integrating OpenAI, handling streaming responses, and multi-provider setups.

Quick Start

Install Molykit in your Makepad project and instantiate a BotContext with a concrete BotClient (for example, OpenAIClient). Then wire the Molykit widgets into your UI and start the event loop to stream responses.

Frequently Asked Questions about molykit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a cross-platform AI chat UI with Makepad?

You can build a cross-platform AI chat UI with Makepad by using Molykit's ready-to-use widgets like Chat, Messages, and PromptInput, combined with async utilities that run natively and in WASM. This removes boilerplate for streaming chat interfaces.

Can I stream OpenAI responses in a Makepad application?

Yes, you can stream OpenAI responses in a Makepad application using Molykit's OpenAI-compatible client with SSE streaming support. It handles streaming flows across native and WASM platforms via async utilities like PlatformSend and spawn.

Does Molykit work with WebAssembly for AI chat interfaces?

Molykit works with WebAssembly for AI chat interfaces by providing cross-platform async utilities such as PlatformSend, spawn, and ThreadToken. These enable async AI chat tasks to run natively and in WASM without changing the core logic.

How do I integrate multiple AI providers into a Makepad chat app?

You integrate multiple AI providers into a Makepad chat app by implementing the BotClient trait for each provider and sharing a BotContext for synchronized UI access. Molykit includes an OpenAIClient and protocol types for multi-provider setups.

What is the best way to handle SSE streaming in Rust cross-platform UIs?

The best way to handle SSE streaming in Rust cross-platform UIs is using a toolkit like Molykit that provides an OpenAI-compatible client with SSE support and async utilities, ensuring streaming responses work across native and WASM environments.