Tsukumo — AI skill for Claude Code
Give your AI agent a body: a 3D VRM desktop companion that becomes the face and voice of Claude Code, Codex, Gemini CLI, OpenClaw or a local LLM.
How to install Tsukumo
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Ananas2916/tsukumo and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Tsukumo does
Give your AI agent a body: a 3D VRM desktop companion that becomes the face and voice of Claude Code, Codex, Gemini CLI, OpenClaw or a local LLM. Local TTS with lip-sync, voice input, offline by default.
Alternatives in AI
- AI Avatar System — 🎭 AI Avatar / digital human platform — upload a photo, clone a voice, talk to any face in real time with lip 422 ★
- Awareness Local — Local-first AI agent memory — one command, works offline, no account needed 199 ★
- Coffee CLI — A clean AI terminal workspace 180 ★
README
Tsukumo
**Give your AI agent a body.** Tsukumo is a 3D desktop companion that becomes the face and voice of whatever brain you plug in: Claude Code, Codex, Gemini CLI, OpenClaw, Antigravity, a local model in LM Studio or Ollama, or a cloud API. It lives on your desktop as a VRM character (or a small spirit flame), talks with lip-synced local text-to-speech, listens when you speak, and taps you on the shoulder when your agents finish their work.
*In Japanese folklore, **tsukumogami** are objects that come alive after a hundred years. Tsukumo does the same to your desk, without the wait.*
[](https://github.com/Ananas2916/tsukumo/releases/latest) [](LICENSE)   

**Download (Windows 10/11):** grab the latest `Tsukumo Setup .exe` from [Releases](https://github.com/Ananas2916/tsukumo/releases/latest). No Python or Node needed. The installer is not code-signed, so SmartScreen asks once: *More info → Run anyway*.
At a glance
| What | Desktop companion: VRM avatar + voice + ears, driven by any LLM or agent |
| Brains | 27 engines: 18 agents (Claude Code, Codex, Gemini CLI, OpenClaw, Antigravity, Cursor CLI, GitHub Copilot CLI, OpenCode, Qwen Code, Amp, Goose, Crush, Factory Droid, Continue CLI, Kiro CLI, Cline, Hermes Agent, any CLI command), 2 local servers (LM Studio / any OpenAI-compatible server, Ollama), 7 cloud APIs (Claude, Gemini, Groq, OpenRouter, DeepSeek, Mistral, Together) |
| Voices | 11 engines: Kokoro (defau |
Related Skills
Fullstack Agent
Give your AI a full stack: memory, voice, face, and hands. This is the "I want an AI agent" shortcut. It sets
Agent Voice
Give any LLM agent a cloned voice: its reply in, one spoken line out, via an OmniVoice TTS server.
10d10s Bonus H3 Lipsync
Claude Code skill - lip-synced AI talking characters with MiniMax H3: script rules H3 can sync, ElevenLabs v4
Zeus
A pixel-art face for your AI assistant. A macOS companion for Claude Code, Codex and OpenCode that asks, shows
Evolve Skill
SPIKE — GEPA/DSPy-style offline A/B evolution of ONE prompt-only skill's SKILL.md body against a small fixture
AI Visualizer
Give your AI agent a face. Four full-screen visualizers. A living circuit board with a cinematic fly-through,
Related Agents
Miband Dev
Builds and installs Xiaomi Smart Band 10 / 10 Pro / 11 watch faces and Vela JS apps (with an Android companion
Trust Reviewer
Traces one retrieved document from the retriever to the context window and names the line where it becomes lab
Abacus AI Agent
Abacus.AI, the company behind ChatLLM and CodeLLM, ships its coding assistant as an agent extension for VS Cod