Auto-sync: 2026-04-28 12:03
This commit is contained in:
@@ -1,41 +1,56 @@
|
||||
---
|
||||
title: "Context Substrate"
|
||||
type: concept
|
||||
tags: [ai-agent, memory, architecture]
|
||||
last_updated: 2026-04-23
|
||||
---
|
||||
|
||||
## Definition
|
||||
AI Agent 的上下文管理技术路线之一(Camp 2)。维护结构化、人类可读的上下文文件(Markdown、知识图谱、上下文容器),跨会话自然累积增长。
|
||||
|
||||
## Core Philosophy
|
||||
**"What context should the AI work inside?"**(而非 Camp 1 的 "what should the AI remember?")
|
||||
|
||||
- Nothing gets extracted — the context is the files.
|
||||
- 文件是人类可读、可编辑、可理解的。
|
||||
- 因为上下文是文件,人可以随时纠正、补充和理解 Agent 知道什么。
|
||||
- 系统随时间自然复合增长(compounding),而非依赖提取质量。
|
||||
|
||||
## Mechanism
|
||||
```
|
||||
Agent reads structured context → Agent works within that context →
|
||||
Agent (or background process) writes back to the structured context →
|
||||
Next session, the context is richer than before
|
||||
```
|
||||
|
||||
## Representative Tools
|
||||
- [[OpenClaw]]:Markdown 文件 + dreaming cycle
|
||||
- [[Zep]]:Temporal knowledge graph(Graphiti)
|
||||
- [[Thoth]]:Personal knowledge graph(10 entity types, 67 relations)
|
||||
- [[TrustGraph]]:Context Cores(可移植版本化上下文捆绑包)
|
||||
- [[MemSearch]]:Markdown-first + shadow vector index
|
||||
- [[ALIVE]]:Structured context substrate, walnuts as portable containers
|
||||
|
||||
## Relationship to Camp 1
|
||||
- Camp 1 优化目标:**召回**(can the system find the right fact?)
|
||||
- Camp 2 优化目标:**复合**(does the system get better over time?)
|
||||
- Zep 从"memory"→"context engineering"的品牌重塑,是 Camp 1/Camp 2 边界处最强的市场信号
|
||||
- Supermemory(Camp 1)的时序感知和 Honcho(Camp 1)的心理建模,代表 Camp 1 向 Camp 2 的演进趋势
|
||||
|
||||
## Key Distinction from RAG
|
||||
RAG 通常指一次性的文档检索问答场景;Context Substrate 强调**跨时间的上下文累积**,是持续运行 Agent 的基础设施。
|
||||
---
|
||||
title: "Context-Substrate"
|
||||
type: concept
|
||||
tags: [AI-Memory, Context-Substrate, File-Native, Compounding]
|
||||
sources: [ai-memory-tools-two-camps]
|
||||
last_updated: 2026-04-15
|
||||
---
|
||||
|
||||
## Definition
|
||||
AI 记忆工具的 Camp 2 范式。维护结构化、人类可读的上下文,跨会话累积。不提取"事实"——上下文本身就是文件。问的核心问题是"**AI 应该在什么上下文中工作?**"
|
||||
|
||||
## Core Loop
|
||||
1. Agent 工作前读取结构化上下文
|
||||
2. Agent 在上下文中工作
|
||||
3. Agent(或后台进程)写回结构化上下文
|
||||
4. 下一会话,上下文比之前更丰富
|
||||
|
||||
## Optimization Goal
|
||||
**复合增长(compounding)**——系统是否随时间变得更好?
|
||||
|
||||
## Representative Tools
|
||||
- [[OpenClaw]]:358k stars,MEMORY.md + 每日笔记 + DREAMS.md,Dreaming 三阶段
|
||||
- [[Zep]]:4.4k stars,从"Memory"重品牌为"Context Engineering",TKG 时间知识图谱
|
||||
- [[Thoth]]:145 stars,10 实体类型 + 67 关系,Dream Cycle 四阶段
|
||||
- [[TrustGraph]]:2.0k stars,Context Cores 版本化容器
|
||||
- [[MemSearch]]:1.2k stars,Markdown 优先,Milvus 影子索引
|
||||
- [[ALIVE]]:文件原生,零依赖,@witcheer 自用方案
|
||||
|
||||
## Common Characteristics
|
||||
- 文件(Markdown、知识图谱)是上下文本身
|
||||
- 人类可读、可编辑、可版本控制
|
||||
- 跨会话累积,不替换而是丰富
|
||||
- 后台整合进程(Dream Cycle)定期提炼
|
||||
- 透明度高:人类能准确知道 Agent 知道什么
|
||||
|
||||
## Key Contrast with Memory-Backend
|
||||
| 维度 | Memory Backend | Context Substrate |
|
||||
|------|---------------|-------------------|
|
||||
| 问的问题 | AI 应该记住什么? | AI 应该在什么上下文中工作? |
|
||||
| 处理对象 | 提取的"事实" | 结构化上下文文件 |
|
||||
| 人类可读 | 否 | 是 |
|
||||
| 随时间演进 | 否(静态条目) | 是(复合累积) |
|
||||
| 透明度 | 低(黑盒) | 高(文件可见) |
|
||||
| 优化目标 | 召回精度 | 复合增长 |
|
||||
| 适用场景 | 单轮问答 | 持续多会话 Agent |
|
||||
|
||||
## Connections
|
||||
- [[Memory-Backend]] ← 对立阵营 ← Context-Substrate
|
||||
- [[OpenClaw]] ← 属于 ← Context-Substrate
|
||||
- [[Zep]] ← 属于 ← Context-Substrate
|
||||
- [[Thoth]] ← 属于 ← Context-Substrate
|
||||
- [[TrustGraph]] ← 属于 ← Context-Substrate
|
||||
- [[MemSearch]] ← 属于 ← Context-Substrate
|
||||
- [[ALIVE]] ← 属于 ← Context-Substrate
|
||||
- [[Context-Engineering]] ← 重品牌化方向 ← Context-Substrate
|
||||
- [[ai-memory-tools-two-camps]] ← 来源 ← Context-Substrate 是 @witcheer 提出的分类框架中的 Camp 2
|
||||
|
||||
Reference in New Issue
Block a user