explaingit

rotejin/tomari-guruguru

Analysis updated 2026-05-18

34JavaScriptAudience · generalComplexity · 2/5LicenseSetup · easy

TLDR

A browser-based animated streaming avatar that follows your mouse and lip-syncs to your microphone for use as an OBS overlay.

Mindmap

mindmap
  root((tomari-guruguru))
    What it does
      Animated streaming avatar
      Mouse and mic reactive
    Tech stack
      JavaScript
      GitHub Pages
    Use cases
      OBS overlay
      Custom characters
    Audience
      Streamers
      VTubers

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Use the default Tomari avatar as a streaming overlay in OBS.

USE CASE 2

Swap in your own character by creating six custom sprite sheets.

USE CASE 3

Adjust mic sensitivity and mouse-follow settings for a chroma-key streaming setup.

USE CASE 4

Host the avatar for free on GitHub Pages as an OBS browser source.

What is it built with?

JavaScriptHTML5GitHub Pages

How does it compare?

rotejin/tomari-guruguruquyanhfex/messenger-recall-toolspanky96/glm-coding-grabber
Stars343434
LanguageJavaScriptJavaScriptJavaScript
Setup difficultyeasymoderatemoderate
Complexity2/53/53/5
Audiencegeneraldeveloperdeveloper

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · easy Time to first run · 5min

Custom characters require creating six sprite sheets and running a slicing script.

The code itself is MIT licensed, but the character artwork has a separate license that restricts commercial use.

In plain English

This is a browser-based animated character avatar built for streaming, specifically designed around a character named Tomari. It runs in a web browser and is intended to be used as an overlay in streaming software such as OBS. The README is written in Japanese. The avatar has two modes. In the first mode, called Guruguru, the character's face turns to follow the mouse cursor across 25 directions (a 5x5 grid of horizontal and vertical angles). In the second mode, called Talk, the character lip-syncs in real time to microphone input or an audio file, with the mouth switching between closed, half-open, and fully open states depending on the volume level. When the character is speaking, the eyes also shift to closed-blinking images. The character images are organized as sprite sheets: six full sheets, each containing a 5x5 grid of poses for one of the six facial states. The app swaps between individual frames depending on where the mouse is pointing and what volume level the audio input is at. A Tweaks panel in the corner of the screen lets you adjust microphone sensitivity, how fast the mouth closes after silence, how closely the character follows the mouse, and the background color. Setting the background to a chroma-key green makes it easy to cut out in OBS. The project is hosted on GitHub Pages. To use a custom character instead of the default Tomari, you need to create six new sprite sheets (one per facial state), run a slicing script included in the repository to cut them into individual frames, and update a configuration file. Instructions for that process are included in the docs folder. The code is MIT licensed, the character art has a separate license that limits commercial use.

Copy-paste prompts

Prompt 1
Walk me through setting up tomari-guruguru as an OBS browser source overlay.
Prompt 2
Explain how the Guruguru and Talk modes work and how to switch between them.
Prompt 3
Show me how to create custom sprite sheets to replace the default Tomari character.
Prompt 4
Help me adjust the Tweaks panel settings for a chroma-key green background.

Frequently asked questions

What is tomari-guruguru?

A browser-based animated streaming avatar that follows your mouse and lip-syncs to your microphone for use as an OBS overlay.

What language is tomari-guruguru written in?

Mainly JavaScript. The stack also includes JavaScript, HTML5, GitHub Pages.

What license does tomari-guruguru use?

The code itself is MIT licensed, but the character artwork has a separate license that restricts commercial use.

How hard is tomari-guruguru to set up?

Setup difficulty is rated easy, with roughly 5min to a first successful run.

Who is tomari-guruguru for?

Mainly general.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.