Show HN: 超速 - 企业 AI 智能体的认知记忆图
1 分•作者: prabhatkr•5 个月前
Superfast 是 Superpowers 智能体框架的演进版本，现已与 FastMemory 集成——一个并发的 Rust 引擎，它将非结构化文本映射到 CBFDAE（组件、块、功能、数据、访问、事件）功能本体中。
虽然 RAG（检索增强生成）已成为 LLM（大型语言模型）“添加知识”的标准方法，但由于语义噪声和分块过程中逻辑边界的破坏，它往往难以大规模应用。 Superfast 将内存视为一个架构层。它利用 Louvain 社区检测算法来数学推导出功能集群，为智能体提供一个跨会话持续存在的确定性“逻辑层”。
我们保留了原始框架严格的 TDD（测试驱动开发）和苏格拉底式的纪律，但将其扩展到 Microsoft Fabric 和 AWS Glue 等环境，在这些环境中，“令牌浪费”是主要的瓶颈。
请在此处查看：<a href="https://github.com/FastBuilderAI/superfast" rel="nofollow">https://github.com/FastBuilderAI/superfast</a>
查看原文
Superfast is an evolution of the Superpowers agent framework, now integrated with FastMemory—a concurrent Rust engine that maps unstructured text into a CBFDAE (Component, Block, Function, Data, Access, Event) functional ontology.<p>While RAG has become the standard for "adding knowledge" to LLMs, it often fails at scale due to semantic noise and the destruction of logical boundaries during chunking. Superfast treats memory as an architectural layer. It utilizes Louvain community detection to mathematically derive functional clusters, giving agents a deterministic "Logic Layer" that persists across sessions.<p>We’ve maintained the strict TDD and Socratic discipline of the original framework but scaled it for environments like Microsoft Fabric and AWS Glue where "token waste" is a primary bottleneck.<p>Check it out here: <a href="https://github.com/FastBuilderAI/superfast" rel="nofollow">https://github.com/FastBuilderAI/superfast</a>