Neo

Active
GitHub JavaScript MIT

Description

The Application Engine for the AI Era. Multi-threaded, AI-native runtime with persistent Scene Graph for real-time agent introspection.

Key Features

  • Cross-model AI engineering team — Claude, Gemini, GPT swarm with shared memory and cross-family review
  • Neural Link possession interface — agents inspect and mutate live application state in real time
  • Native Edge Graph with DreamService for REM-cycle consolidation that re-steers priorities from friction
  • Multi-threaded application engine — App Worker, VDom Worker, Data Worker, Canvas Worker with zero-build ES modules
  • Agent OS with Memory Core, Knowledge Base, A2A coordination, and GitHub workflow automation
  • Self-evolving MX loop — friction becomes tickets, PRs, skills, memory, and improved graph topology

Use Cases

💡 Running a self-evolving AI engineering team on your own codebase with persistent memory
💡 Using cross-model review to catch issues no single model can see in isolation
💡 Building production multi-threaded web applications with AI-native runtime
💡 Deploying the Agent OS as a multi-tenant cloud service for multiple codebases
💡 Leveraging DreamService cycles to consolidate learnings and improve agent performance over time

Tags

runtime ai-native multithreading javascript agent-memory

Categories

🤖 Agent Framework 🧩 Agent Memory

Quick Start

1. Clone the repo: `git clone https://github.com/neomjs/neo.git`
2. Install dependencies: `npm install`
3. Explore the AI module at `/ai/` — Memory Core, Knowledge Base, and Native Edge Graph
4. Read the Architecture Overview at `learn/benefits/ArchitectureOverview.md`
5. For cloud deployment, see `learn/agentos/cloud-deployment/Day0Tutorial.md`
Visit GitHub

Related Projects