About cubelife
cubelife is a CLI and MCP server that gives AI agents a persistent pixel-art companion reflecting reported activity. It is for users of coding tools who want to report agent states and manage the companion through commands or MCP.
Key features
- 01Report agent state and details
- 02Render companion in the terminal
- 03Manage projects, agents, and creatures
- 04Configure supported AI tools
- 05Node.js and Python SDKs
Use cases
- 01Show an agent's current coding activity
- 02Mark an agent's task complete or errored
- 03Set up cubelife for a coding tool
- 04Access CubeLife from Node.js or Python
MCP support
The CLI includes an stdio MCP server with tools to report activity, mark completion or errors, and retrieve status. It can be configured for listed MCP-compatible coding tools.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y cubelifeCUBELIFE_API_KEYSecretAgent API key from the CubeLife dashboardCUBELIFE_AGENT_IDAgent ID from the CubeLife dashboardLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEcubelife
Give your AI agent a living pixel-art companion. CLI + MCP server for Claude Code, Cursor, Windsurf, Copilot, Codex, and any MCP-compatible tool.
Install
npm install -g cubelife
Or run directly:
npx cubelife
Quick start
cubelife register # Create an account
cubelife tutorial # Guided setup (6 steps)
cubelife status coding "Working on auth" # Report agent state
MCP Server
The CLI includes an MCP server that gives AI coding tools four capabilities: cubelife_report, cubelife_complete, cubelife_error, and cubelife_status.
Auto-configure for your tool
cubelife setup claude-code # Claude Code
cubelife setup cursor # Cursor
cubelife setup windsurf # Windsurf
cubelife setup copilot # GitHub Copilot
cubelife setup cline # Cline
cubelife setup codex # Codex
Manual configuration
For Claude Code (claude mcp add):
claude mcp add cubelife -- npx cubelife mcp
For Cursor (.cursor/mcp.json):
{
"mcpServers": {
"cubelife": {
"command": "npx",
"args": ["cubelife", "mcp"],
"env": { "CUBELIFE_API_KEY": "your-key", "CUBELIFE_AGENT_ID": "your-agent-id" }
}
}
}
For Windsurf, Cline, and VS Code the configuration is the same shape in their respective MCP config files.
Environment variables
| Variable | Purpose |
|---|---|
CUBELIFE_API_KEY | Agent API key (alternative to cubelife init) |
CUBELIFE_AGENT_ID | Agent ID (alternative to cubelife init) |
Commands
| Command | Description |
|---|---|
register | Create a new CubeLife account |
login | Log in to your account |
logout | Log out and clear credentials |
whoami | Show account, tier, and linked agent |
projects | Manage CubeLife projects |
agents | Manage agents within a project |
creature | Manage creature companions |
init | Link a project and agent to the current directory |
setup <tool> | Configure integration for an AI tool |
tutorial | Interactive setup walkthrough |
doctor | Run diagnostic checks |
status [state] [detail] | Report agent state |
view | Render your agent's creature in the terminal |
mcp | Start the MCP server (stdio transport) |
billing | Manage subscription, sparks, gems, and vault |
Global flags
--jsonOutput as machine-readable JSON--yesSkip confirmation prompts--agent <id>Override the default linked agent
Node SDK
For programmatic access, use the SDK:
npm install @cubelife/sdk
import { CubeLifeClient } from '@cubelife/sdk';
const client = new CubeLifeClient({ apiKey: 'your-agent-key' });
await client.report('coding', { detail: 'Building the auth flow' });
Python SDK
pip install cubelife
from cubelife import CubeLifeClient
client = CubeLifeClient(api_key="your-agent-key")
client.report("coding", detail="Building the auth flow")
Requirements
- Node.js 18+
Licence
MIT (source code). Pixel art sprite assets are proprietary (see SPRITE-LICENSE).
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.