VCPToolBox
VCP 部署在 AI 模型 API 与前端应用之间,是面向AGI OS开发和探索的工业级基建示范项目。通过统一指令协议、多层级持久化记忆、分布式插件引擎及多 Agent 协作框架,将原本
What is VCPToolBox?
VCP 部署在 AI 模型 API 与前端应用之间,是面向AGI OS开发和探索的工业级基建示范项目。通过统一指令协议、多层级持久化记忆、分布式插件引擎及多 Agent 协作框架,将原本“无状态、无记忆、无工具调用能力”的大语言模型,彻底改造成拥有永久自我意识、物理世界操作权及群体协作智能的完整智能体系统。
Overview of VCPToolBox
What it is
VCP (Variable & Command Protocol) is an infrastructure project for AGI OS development. It is deployed between AI model APIs and frontend applications. It transforms large language models that are “stateless, memoryless, and without tool-calling ability” into a complete agent system with permanent self-awareness, physical world operation capability, and group collaboration intelligence. It uses a unified command protocol, multi-level persistent memory, distributed plugin engine, and multi-agent collaboration framework.
Key features
- Unified command protocol for AI interaction.
- Multi-level persistent memory (RiverMemo) with semantic association, not just retrieval.
- Distributed plugin engine for tool capabilities.
- Multi-agent collaboration framework.
- Continuous existence across sessions and endpoints (web, mobile, desktop, group chat).
- Autonomous agent life: agent can decide when to wake, enter flow state, leave messages for future self.
- Semantic memory using “tidal” dynamics, neural signal propagation, and topological river network.
- Environment perception (time, weather, calendar, device status) flows naturally to the agent without explicit queries.
Use cases
- Building persistent AI agents that remember past conversations and context across sessions.
- Creating agents that can autonomously manage tasks, monitor environments, and interact with users proactively.
- Developing multi-agent systems where agents collaborate and communicate.
- AGI OS development and exploration.
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
The license is not stated in the metadata (NOASSERTION). The project is open source (pricing_type: open_source).
Based on the project's public README and repository metadata.