memory-integration

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Memory Integration

记忆集成

Overview

概述

Use both memory systems to maintain context across sessions.

Core principle: You have no memory between sessions. Use these tools to remember.

Systems:

Episodic Memory - Conversation history search
Knowledge Graph (mcp__memory) - Structured facts and relationships

同时使用两种记忆系统来在会话间维持上下文。

核心原则： 会话之间没有默认记忆，需使用这些工具来留存信息。

系统组成：

Episodic Memory - 对话历史检索
知识图谱（mcp__memory）- 结构化事实与关系

When to Use Memory

何时使用记忆功能

Moment	Memory Action
Session start	Search for relevant context
Before starting issue	Search for previous work
Making decision	Check for past decisions
Completing work	Store important learnings
Session end	Store key outcomes

场景	记忆操作
会话开始	检索相关上下文
开始处理问题前	检索过往工作记录
做决策时	查看过往决策
完成工作后	存储重要经验
会话结束	存储关键成果

Episodic Memory

What It Stores

存储内容

Full conversation history
Decisions made
Problems solved
Approaches tried
Lessons learned

完整对话历史
已做出的决策
已解决的问题
尝试过的方案
积累的经验

Searching Episodic Memory

检索Episodic Memory

Use the episodic-memory skill or MCP tools:

Search for:
- Issue number: "issue 123", "#123"
- Feature name: "authentication", "user login"
- Problem type: "TypeScript error", "build failure"
- Project name: repository name

使用episodic-memory技能或MCP工具：

检索关键词：
- 问题编号："issue 123", "#123"
- 功能名称："authentication", "user login"
- 问题类型："TypeScript error", "build failure"
- 项目名称：仓库名称

Semantic Search (Single Query)

语义检索（单查询）

javascript

// Search with natural language
mcp__plugin_episodic-memory_episodic-memory__search({
  query: "user authentication implementation decisions"
})

javascript

// 使用自然语言检索
mcp__plugin_episodic-memory_episodic-memory__search({
  query: "user authentication implementation decisions"
})

Precise Search (Multiple Concepts)

精准检索（多概念）

javascript

// Search for intersection of concepts
mcp__plugin_episodic-memory_episodic-memory__search({
  query: ["authentication", "session", "JWT"]
})

javascript

// 检索多个概念的交集
mcp__plugin_episodic-memory_episodic-memory__search({
  query: ["authentication", "session", "JWT"]
})

Reading Full Conversations

读取完整对话

After finding relevant results:

javascript

// Read the full conversation
mcp__plugin_episodic-memory_episodic-memory__read({
  path: "/path/to/conversation.jsonl"
})

找到相关结果后：

javascript

// 读取完整对话
mcp__plugin_episodic-memory_episodic-memory__read({
  path: "/path/to/conversation.jsonl"
})

What to Search For

检索方向

Situation	Search Terms
Starting issue #123	"issue 123", "#123"
Working on auth	"authentication", "login", "session"
TypeScript problem	"TypeScript", "type error", specific error message
Similar feature	Feature name, related concepts

场景	检索关键词
开始处理issue #123	"issue 123", "#123"
处理认证功能	"authentication", "login", "session"
TypeScript问题	"TypeScript", "type error", 具体错误信息
类似功能	功能名称、相关概念

Knowledge Graph (mcp__memory)

知识图谱（mcp__memory）

What It Stores

存储内容

Entities (Projects, Issues, Decisions, Patterns)
Relationships between entities
Observations about entities

实体（项目、问题、决策、模式）
实体间的关系
实体相关的观察记录

Creating Entities

创建实体

Store important facts:

javascript

// Create an entity for a project decision
mcp__memory__create_entities({
  entities: [{
    name: "Decision: Use JWT for Auth",
    entityType: "Decision",
    observations: [
      "Decided on 2024-12-01",
      "JWT chosen over sessions for API statelessness",
      "Related to issue #123",
      "Implementation in src/auth/jwt.ts"
    ]
  }]
})

存储重要事实：

javascript

// 为项目决策创建实体
mcp__memory__create_entities({
  entities: [{
    name: "Decision: Use JWT for Auth",
    entityType: "Decision",
    observations: [
      "Decided on 2024-12-01",
      "JWT chosen over sessions for API statelessness",
      "Related to issue #123",
      "Implementation in src/auth/jwt.ts"
    ]
  }]
})

Creating Relationships

创建关系

Link entities together:

javascript

// Create relationships
mcp__memory__create_relations({
  relations: [
    {
      from: "Project: MyApp",
      to: "Decision: Use JWT for Auth",
      relationType: "has_decision"
    },
    {
      from: "Issue #123",
      to: "Decision: Use JWT for Auth",
      relationType: "resulted_in"
    }
  ]
})

关联不同实体：

javascript

// 创建关系
mcp__memory__create_relations({
  relations: [
    {
      from: "Project: MyApp",
      to: "Decision: Use JWT for Auth",
      relationType: "has_decision"
    },
    {
      from: "Issue #123",
      to: "Decision: Use JWT for Auth",
      relationType: "resulted_in"
    }
  ]
})

Searching the Graph

检索图谱

javascript

// Search for relevant nodes
mcp__memory__search_nodes({
  query: "authentication"
})

// Open specific nodes
mcp__memory__open_nodes({
  names: ["Decision: Use JWT for Auth"]
})

// Read entire graph (for small graphs)
mcp__memory__read_graph({})

javascript

// 检索相关节点
mcp__memory__search_nodes({
  query: "authentication"
})

// 打开特定节点
mcp__memory__open_nodes({
  names: ["Decision: Use JWT for Auth"]
})

// 读取完整图谱（适用于小型图谱）
mcp__memory__read_graph({})

Memory Protocol

记忆协议

At Session Start

会话开始时

Search episodic memory for:
- Current issue number
- Project/repository name
- Active feature being worked on
Search knowledge graph for:
- Project entity
- Related decisions
- Known patterns
Synthesize context before proceeding

检索Episodic Memory：
- 当前问题编号
- 项目/仓库名称
- 正在开发的功能
检索知识图谱：
- 项目实体
- 相关决策
- 已知模式
合成上下文后再开展工作

During Work

工作过程中

Store as you go:

Event	Store In
Major decision	Knowledge graph entity
Problem solved	Add observation to issue entity
Pattern discovered	Knowledge graph entity
Lesson learned	Add observation

随时存储信息：

事件	存储位置
重大决策	知识图谱实体
问题解决	为问题实体添加观察记录
发现模式	知识图谱实体
积累经验	添加观察记录

At Session End

会话结束时

Update knowledge graph with:
- New decisions made
- Problems solved
- Patterns discovered
Add observations to existing entities:
- Progress on issues
- Learnings
- Next steps

更新知识图谱：
- 新增决策
- 已解决的问题
- 发现的新模式
为现有实体添加观察记录：
- 问题进展
- 积累的经验
- 下一步计划

Entity Types

实体类型

Suggested entity types for the knowledge graph:

Type	Use For	Example
Project	Repository/codebase	"Project: MyApp"
Issue	GitHub issues	"Issue #123: Auth"
Decision	Architectural decisions	"Decision: Use JWT"
Pattern	Code patterns	"Pattern: Repository Layer"
Problem	Known issues	"Problem: Race Condition in X"
Person	Collaborators	"Person: Alice (maintainer)"

知识图谱建议使用的实体类型：

类型	适用场景	示例
Project	仓库/代码库	"Project: MyApp"
Issue	GitHub问题	"Issue #123: Auth"
Decision	架构决策	"Decision: Use JWT"
Pattern	代码模式	"Pattern: Repository Layer"
Problem	已知问题	"Problem: Race Condition in X"
Person	协作者	"Person: Alice (maintainer)"

Example: Issue Memory Flow

示例：问题记忆流程

Session 1: Starting Issue

会话1：开始处理问题

javascript

// Search for any previous context
const episodic = await search("issue 456 user profile");
const graph = await search_nodes("user profile");

// If nothing found, create fresh entity
await create_entities({
  entities: [{
    name: "Issue #456: User Profile Page",
    entityType: "Issue",
    observations: [
      "Started: 2024-12-01",
      "Scope: Profile display, edit, avatar upload"
    ]
  }]
});

javascript

// 检索过往上下文
const episodic = await search("issue 456 user profile");
const graph = await search_nodes("user profile");

// 若无结果，创建新实体
await create_entities({
  entities: [{
    name: "Issue #456: User Profile Page",
    entityType: "Issue",
    observations: [
      "Started: 2024-12-01",
      "Scope: Profile display, edit, avatar upload"
    ]
  }]
});

Session 1: Mid-Work Decision

会话1：工作中做出决策

javascript

// Store a decision made
await add_observations({
  observations: [{
    entityName: "Issue #456: User Profile Page",
    contents: [
      "Decision: Using react-image-crop for avatar cropping",
      "Reason: Best mobile support, active maintenance"
    ]
  }]
});

javascript

// 存储已做出的决策
await add_observations({
  observations: [{
    entityName: "Issue #456: User Profile Page",
    contents: [
      "Decision: Using react-image-crop for avatar cropping",
      "Reason: Best mobile support, active maintenance"
    ]
  }]
});

Session 2: Resuming

会话2：继续处理

javascript

// Search for context
const results = await search("issue 456");
// Read: "Using react-image-crop for avatar cropping"

// Continue with context maintained

javascript

// 检索上下文
const results = await search("issue 456");
// 读取到："Using react-image-crop for avatar cropping"

// 基于留存的上下文继续工作

What to Store

存储建议

Always Store

需存储的内容

Architectural decisions with rationale
Non-obvious problem solutions
Important constraints discovered
Dependencies between components

带有决策依据的架构决策
非显而易见的问题解决方案
发现的重要约束
组件间的依赖关系

Don't Store

无需存储的内容

Trivial implementation details
Things obvious from code
Temporary debugging notes
Speculation without conclusion

琐碎的实现细节
代码中显而易见的内容
临时调试笔记
无结论的推测

Checklist

检查清单

Integration

集成说明

This skill is called by:

```
session-start
```
- Initial context gathering
```
issue-driven-development
```
- Step 4

This skill supports:

Cross-session continuity
Decision documentation
Pattern discovery

本技能被以下功能调用：

```
session-start
```
- 初始上下文收集
```
issue-driven-development
```
- 步骤4

本技能支持：

跨会话连续性
决策文档化
模式发现