Skip to content

GeminiLight/evo-agent-team

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

33 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ–₯️ evo-agent-team

npm version License Node.js

A real-time dashboard for monitoring and visualizing Claude Code agent teams β€” rendered in a retro CRT/phosphor terminal aesthetic.

Features β€’ Demo β€’ Installation β€’ Documentation β€’ Contributing


πŸ“Έ Demo

Dashboard Preview

Real-time agent team visualization with CRT terminal aesthetic


✨ Features

Feature Description
βš™οΈ Approval System Permission request approval panel, global modal, toast notifications, browser alerts
πŸ›‘οΈ Supervision Rules Rule-based agent supervision with auto-extraction from feedback
πŸŽ“ Onboarding Tour 5-step interactive guided walkthrough for new users
πŸ“Š Matrix View Dashboard with team overview, agent roster cards, and filterable task registry
πŸ•ΈοΈ Graph View Interactive DAG topology of agents and task dependencies powered by ReactFlow
πŸ’¬ Comms View Agent communication log with per-agent filtering and message type badges
πŸ“… Timeline View Chronological feed of all task status changes with transition history
πŸ” Task Detail Panel Click any task to open a sliding panel with full description, owner, and dependency info
πŸ“ Review Panel Feedback log, agent preferences management, and statistics with LLM-powered insights
🧠 Memory Management View and edit team MEMORY.md with AI-powered memory extraction
πŸ”„ Knowledge Transfer Cross-team knowledge migration with LLM analysis and classification
πŸ“‹ Context Summary Three-section context (Decisions/Progress/Context) with token budget
🎯 Exec Summary AI-generated team progress with caching
πŸ“€ Export Download the topology graph as PNG or the full team state as JSON
⚑ Real-time Sync WebSocket push updates with automatic polling fallback
🎨 12 Themes 7 dark + 5 light terminal-inspired themes
🌍 i18n Full English/Chinese bilingual support
🎭 Demo Mode Built-in mock team auto-activates when no live teams detected

πŸš€ Quick Start

Prerequisites

  • Node.js 18+
  • Claude Code (optional β€” demo mode works standalone)

Installation

# Install globally
npm install -g evo-agent-team

# Run the dashboard
evo-agent-team

Or use without installing:

npx evo-agent-team

Then open http://localhost:5173


πŸ“– Documentation

Usage

# Start with default settings
evo-agent-team

# With custom port (server)
PORT=4000 evo-agent-team

# Disable demo mode
DEMO_MODE=off evo-agent-team

Configuration

Create .env file in your project root:

# Data directories
TEAMS_DIR=~/.claude/teams
TASKS_DIR=~/.claude/tasks

# Server settings
PORT=3006
POLL_INTERVAL_MS=2000
DEMO_MODE=auto

# LLM configuration (for AI features β€” supports 20+ providers)
LLM_PROVIDER=openai-compatible
LLM_API_KEY=your-api-key
LLM_BASE_URL=https://api.openai.com/v1
LLM_MODEL=claude-sonnet-4-6
Variable Default Description
TEAMS_DIR ~/.claude/teams Claude Code team configs
TASKS_DIR ~/.claude/tasks Claude Code task files
PORT 3006 API server port
DEMO_MODE auto auto/on/off
LLM_PROVIDER openai-compatible Provider: openai, anthropic, google, groq, openai-compatible
LLM_API_KEY - Required for AI features (summaries, extraction, supervision)
LLM_BASE_URL - Custom API endpoint (for openai-compatible provider)
LLM_MODEL - Model identifier

Data Sources

The dashboard reads from:

Source Path
Team configs ~/.claude/teams/{team}/config.json
Task files ~/.claude/tasks/{team}/{task-id}.json
Agent inboxes ~/.claude/teams/{team}/inboxes/{agent}.json
Memory ~/.claude-internal/projects/{cwd}/memory/MEMORY.md
Feedback ~/.claude/teams/{team}/feedback/{agent}.jsonl
Supervision rules ~/.claude/teams/{team}/TEAM_GUIDE.md ## Supervision Rules

πŸ—οΈ Architecture

Tech Stack

Layer Technology
Frontend React 18, TypeScript, Vite, Tailwind CSS v4
State React Hooks, WebSocket
Graph @xyflow/react (ReactFlow)
i18n react-i18next
Backend Node.js, Express 4, TypeScript ESM
Real-time WebSocket (ws), fs.watch
AI Multi-provider LLM via @mariozechner/pi-ai (OpenAI, Anthropic, Google, Groq)

Project Structure

agent-team/
β”œβ”€β”€ server/               # Express API (TypeScript ESM)
β”‚   └── src/
β”‚       β”œβ”€β”€ routes/       # API endpoints
β”‚       β”œβ”€β”€ index.ts      # Server entry
β”‚       └── config.ts     # Environment config
β”œβ”€β”€ client/               # React 18 + Vite
β”‚   └── src/
β”‚       β”œβ”€β”€ components/   # UI components
β”‚       β”œβ”€β”€ hooks/        # Custom hooks
β”‚       β”œβ”€β”€ utils/        # Utilities
β”‚       └── i18n.ts       # Translations
└── wiki/                 # Documentation
    β”œβ”€β”€ 01-project-roadmap.md
    β”œβ”€β”€ 02-system-architecture.md
    └── 90-changelog.md

πŸ› οΈ Development

# Clone repository
git clone <repo-url>
cd agent-team

# Install dependencies
npm install

# Start development
npm run dev

# Build for production
npm run build

# Start production server
npm run start -w server

Scripts

Script Description
npm run dev Start client + server in dev mode
npm run build Build client for production
npm run start -w server Start production server

🎨 Design System

CRT phosphor terminal aesthetic:

  • Phosphor (#39ff6a) β€” primary accent
  • Amber (#f5a623) β€” warnings/in-progress
  • Crimson (#ff3b5c) β€” errors/blocked
  • Ice (#7eb8f7) β€” informational
  • Void (#040608) β€” background
  • Fonts: JetBrains Mono + Syne

🀝 Contributing

Contributions are welcome! Please read our Contributing Guide for details.

  1. Fork the repository
  2. Create your feature branch (git checkout -b feature/amazing-feature)
  3. Commit your changes (git commit -m 'Add amazing feature')
  4. Push to the branch (git push origin feature/amazing-feature)
  5. Open a Pull Request

πŸ“ License

MIT Β© 2026


πŸ”— Links

About

No description, website, or topics provided.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages