receptron

MulmoChat — Communication skill for Claude Code

Communication community

Research prototype for chat that is not a text stream: keep talking while images, maps and playable games appear on the canvas.

How to install MulmoChat

This entry records only its repository, not the path inside it, so there is no exact command to give. Open receptron/MulmoChat and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What MulmoChat does

Research prototype for chat that is not a text stream: keep talking while images, maps and playable games appear on the canvas. Each kind of answer is a plugin on the GUI Chat Protocol.

Alternatives in Communication

YouTube Shorts Research: Claude SEO — YouTube Shorts Research: Claude SEO skill 5k ★
Codepilot — A native desktop GUI for Claude Code — chat, code, and manage projects visually 4.3k ★
X Research Skill — X/Twitter research skill for Claude Code and OpenClaw 974 ★

README

MulmoChat

**MulmoChat is a research prototype exploring a new paradigm for multimodal AI chat experiences.**

Traditional chat interfaces are fundamentally text-based—users interact through messages and receive responses as text. MulmoChat reimagines this interaction model: users engage in natural conversation while simultaneously experiencing rich visual and interactive content directly on canvas.

The key insight: **AI conversations don't have to be limited to text streams.** By designing an architecture where visual experiences and language understanding coexist naturally, we enable a fundamentally different way of interacting with AI—one where images materialize, maps become explorable, games turn playable, all within the conversational flow.

MulmoChat demonstrates the architecture, design patterns, and user experience principles necessary to build truly multimodal chat interfaces where visual and textual communication work together seamlessly.

Documentation

LLM_OS.md Who should read: product strategists and designers exploring the AI-native OS mindset. Start here for the high-level intent paradigm; jump to WHITEPAPER.md when you need architecture specifics.
WHITEPAPER.md Who should read: engineers and researchers implementing or evaluating the orchestration stack. Dive in for system diagrams and workflow detail; reference LLM_OS.md for the broader narrative framing.
TOOLPLUGIN.md Who should read: developers extending MulmoChat with new capabilities. Follow this to implement plugins end-to-end, from TypeScript contracts to Vue views and configuration.

Getting Started

Install dependencies:

yarn install

Create .env file with following API keys:

OPENAI_API_KEY=...
GEMINI_API_KEY=...
GOOGLE_MAP_API_KEY=... (optional, required for map features)
EXA_API_KEY=... (optional, required for AI-powered search)
ANTHROPIC_API_KEY=... (optional, require