GH Repository · lioensky
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。
- stars
- 2,302
- 30-day movement
- starts with the next reading
- Related entries
- 60
- Connections
- 2
python/poetrynodevuedockervector-databaserustragagent-frameworkprompt-engineeringai-agentai-assistantcontext-management-systemai-companionJavaScriptcontext-managementpythonllmfunction-callingmulti-modelnodejsopenai-compatibleplugin-system
VCPToolBox is a middleware framework deployed between LLM APIs and frontend applications, described as an infrastructure demonstration for AGI OS development. It adds persistent memory, a plugin engine, unified instruction protocols, and multi-agent collaboration to stateless language models.
Use it when you want to give an OpenAI-compatible LLM stack tool calling, persistent context, and multi-agent coordination without rebuilding that plumbing yourself.
Use it to
- Add layered persistent memory to a stateless LLM application
- Run a distributed plugin system for tool invocation
- Coordinate multiple agents in one framework
- Bridge OpenAI-compatible APIs with custom frontends
- Prototype context management with vector database support
For Developers building agent systems on top of LLM APIs
- Role
- agent-framework
- Language
- JavaScript
- Licence
- custom (licence file present)
- Forks
- 377
- Last push
- 2026-09-16
- Latest release
- VCP · 2026-03-12
topicsagent-frameworkllmplugin-systemcontext-managementmemorymiddleware