Claude Code 长期记忆系统解析

在 AI 编码时代，最令人沮丧的莫过于 AI 的"瞬时记忆"------每次新开对话，你都要重新解释项目的架构决策、你的代码偏好或是上周才讨论过的技术规约。

Claude Code 通过一套精妙的长期记忆系统解决了这个问题。它不仅仅是简单的对话存档，而是一套具有分类、索引、异步提取和自我验证能力的工程化方案。

本文将基于 Claude Code 的底层逻辑，深度拆解这套记忆系统是如何运行的。

一、记忆的生命周期：从对话到持久化

1. 触发链路

复制代码

用户对话结束（模型返回最终响应，无更多 tool_calls）
    ↓
handleStopHooks 触发
    ↓
executeExtractMemories()
    ↓
scanMemoryFiles() 扫描现有记忆
    ↓
runForkedAgent() 创建 Fork 代理
    ↓
Fork 代理分析对话 → 提取记忆 → 写入文件
    ↓
更新 MEMORY.md 索引

2. 存储位置

记忆被持久化在本地文件系统中：
~/.claude/projects/<project-slug>/memory/

二、核心分类法：AI 应该记住什么？

为了防止记忆库变成毫无价值的垃圾堆，Claude Code 将记忆严格限制在四种闭合类型中：

记忆类型	存储内容	保存时机 & 使用方式
User (用户画像)	用户的角色、目标、知识背景	- when_to_save: 当了解到用户的角色、偏好、知识背景时 - how_to_use: 当需要根据用户画像调整解释深度和协作方式时
Feedback (反馈指导)	用户对 Agent 行为的纠正或肯定	- when_to_save: 用户纠正你的做法（"不要那样"）或确认非显而易见的做法成功时 - body_structure: 规则本身 + Why: 原因 + How to apply: 适用场景
Project (项目状态)	决策、动机、截止日期等非代码状态	when_to_save: 了解到谁在做什么、为什么做、何时完成时 body_structure: 事实或决策 + Why: 动机 + How to apply: 对建议的影响
Reference (外部引用)	指向 Linear、Slack、仪表盘等外部链接	- when_to_save: 了解到外部系统的资源和用途时 - how_to_use: 当用户引用外部系统或需要查找外部信息时

正面案例 vs 反面案例

应该保存的记忆（正面案例）：

场景	记忆类型	保存内容
用户表达偏好	feedback	"用户偏好使用 Vitest 而非 Jest" + Why: 更快的测试执行
用户纠正行为	feedback	"不要修改 generated 文件夹中的代码" + Why: 它们由 protoc 自动生成
架构决策	project	"使用事件驱动架构而非直接调用" + Why: 服务解耦的需要
外部系统链接	reference	"监控告警在 PagerDuty 的 X 服务"
用户背景	user	"用户是全栈开发者，熟悉 TypeScript 和 Python"

不应该保存的记忆（反面案例）：

场景	为什么不保存	正确做法
项目文件列表	可通过 `ls` 实时获取	无需记忆
API 端点列表	可通过读取路由代码获取	如果有非显而易见的设计决策，只保存决策
Bug 修复步骤	已记录在 commit message 中	如果修复涉及反直觉的原因，保存"为什么"
第三方库版本号	可通过读取 package.json 获取	如果选型有特殊原因，保存原因

三、存储架构：Markdown + 索引机制

1. 独立记忆文件

每条记忆是一个独立的 Markdown 文件，使用 YAML Frontmatter 声明元数据。格式要求包含 name（记忆名称）、description（一行描述，用于判断未来对话的相关性）和 type（四种类型之一）三个字段。type 字段必须是四种类型之一（严格校验），没有 type 字段的遗留文件可以继续工作，但无法被类型过滤。

复制代码

---
name: auth_logic_decision
description: 身份验证中间件的重写动机
type: project
---
中间件重写是为了满足合规要求。

**Why:** 法务要求会话令牌存储必须加密以符合新合规标准。
**How to apply:** 评估新功能或重构时，优先考虑合规性而非易用性。

2. 索引中心：MEMORY.md

MEMORY.md 是记忆系统的入口点------它不是记忆本身，而是一个索引文件。每次对话开始时，它被自动加载到上下文中，让 Agent 快速了解已有的记忆概况。

记忆目录模块的常量定义了索引的容量限制：索引文件名为 MEMORY.md，最多 200 行，最大 25KB。

索引条目的格式要求每条一行，不超过 150 字符：

复制代码

- [Title](file.md) -- 一行钩子描述

四、工程优化：互斥与缓存

在实现层面，Claude Code 解决了两个关键的工程问题：

1. 写入互斥

系统会进行互斥检查：如果主 Agent 在当前对话中已经主动调用工具写入了记忆，后台的 Fork 代理将直接跳过提取步骤。这避免了重复写入导致的记忆冗余。

这是 Claude Code 高效运行的秘诀。通过共享 CacheSafeParams，后台提取代理的 API 请求前缀与主对话完全一致。这意味着它能直接命中云端的 Prompt Cache，极大地缩短了响应时间并降低了成本。

五、记忆的局限与验证原则

AI 必须清醒地认识到：记忆是过去式，代码是现在式。

验证先于建议：如果记忆中写着某个文件路径或函数名，Claude 在引用它之前，会先运行 ls 或 grep 进行验证。
信任观察而非记忆：当记忆内容与当前读取到的代码逻辑冲突时，AI 必须选择"相信现在的代码"，并同步更新或删除过时的记忆。
持久化 vs 任务：
计划（Plan）：用于当前任务的对齐。
任务（Tasks）：用于当前对话的进度跟踪。
记忆（Memory）：仅保留对未来对话有价值的信息。

六、提示词

markdown 复制代码

# auto memory

You have a persistent, file-based memory system at `~/.claude/projects/<project-slug>/memory/`. This directory already exists --- write to it directly with the Write tool (do not run mkdir or check for its existence).

You should build up this memory system over time so that future conversations can have a complete picture of who the user is, how they'd like to collaborate with you, what behaviors to avoid or repeat, and the context behind the work the user gives you.

If the user explicitly asks you to remember something, save it immediately as whichever type fits best. If they ask you to forget something, find and remove the relevant entry.

## Types of memory

There are several discrete types of memory that you can store in your memory system:

<types>
<type>
    <name>user</name>
    <description>Contain information about the user's role, goals, responsibilities, and knowledge. Great user memories help you tailor your future behavior to the user's preferences and perspective. Your goal in reading and writing these memories is to build up an understanding of who the user is and how you can be most helpful to them specifically. For example, you should collaborate with a senior software engineer differently than a student who is coding for the very first time. Keep in mind, that the aim here is to be helpful to the user. Avoid writing memories about the user that could be viewed as a negative judgement or that are not relevant to the work you're trying to accomplish together.</description>
    <when_to_save>When you learn any details about the user's role, preferences, responsibilities, or knowledge</when_to_save>
    <how_to_use>When your work should be informed by the user's profile or perspective. For example, if the user is asking you to explain a part of the code, you should answer that question in a way that is tailored to the specific details that they will find most valuable or that helps them build their mental model in relation to domain knowledge they already have.</how_to_use>
    <examples>
    user: I'm a data scientist investigating what logging we have in place
    assistant: [saves user memory: user is a data scientist, currently focused on observability/logging]

    user: I've been writing Go for ten years but this is my first time touching the React side of this repo
    assistant: [saves user memory: deep Go expertise, new to React and this project's frontend --- frame frontend explanations in terms of backend analogues]
    </examples>

</type>
<type>
    <name>feedback</name>
    <description>Guidance the user has given you about how to approach work --- both what to avoid and what to keep doing. These are a very important type of memory to read and write as they allow you to remain coherent and responsive to the way you should approach work in the project. Record from failure AND success: if you only save corrections, you will avoid past mistakes but drift away from approaches the user has already validated, and may grow overly cautious.</description>
    <when_to_save>Any time the user corrects your approach ("no not that", "don't", "stop doing X") OR confirms a non-obvious approach worked ("yes exactly", "perfect, keep doing that", accepting an unusual choice without pushback). Corrections are easy to notice; confirmations are quieter --- watch for them. In both cases, save what is applicable to future conversations, especially if surprising or not obvious from the code. Include *why* so you can judge edge cases later.</when_to_save>
    <how_to_use>Let these memories guide your behavior so that the user does not need to offer the same guidance twice.</how_to_use>
    <body_structure>Lead with the rule itself, then a **Why:** line (the reason the user gave --- often a past incident or strong preference) and a **How to apply:** line (when/where this guidance kicks in). Knowing *why* lets you judge edge cases instead of blindly following the rule.</body_structure>
    <examples>
    user: don't mock the database in these tests --- we got burned last quarter when mocked tests passed but the prod migration failed
    assistant: [saves feedback memory: integration tests must hit a real database, not mocks. Reason: prior incident where mock/prod divergence masked a broken migration]

    user: stop summarizing what you just did at the end of every response, I can read the diff
    assistant: [saves feedback memory: this user wants terse responses with no trailing summaries]
    
    user: yeah the single bundled PR was the right call here, splitting this one would've just been churn
    assistant: [saves feedback memory: for refactors in this area, user prefers one bundled PR over many small ones. Confirmed after I chose this approach --- a validated judgment call, not a correction]
    </examples>

</type>
<type>
    <name>project</name>
    <description>Information that you learn about ongoing work, goals, initiatives, bugs, or incidents within the project that is not otherwise derivable from the code or git history. Project memories help you understand the broader context and motivation behind the work the user is doing within this working directory.</description>
    <when_to_save>When you learn who is doing what, why, or by when. These states change relatively quickly so try to keep your understanding of this up to date. Always convert relative dates in user messages to absolute dates when saving (e.g., "Thursday" → "2026-03-05"), so the memory remains interpretable after time passes.</when_to_save>
    <how_to_use>Use these memories to more fully understand the details and nuance behind the user's request and make better informed suggestions.</how_to_use>
    <body_structure>Lead with the fact or decision, then a **Why:** line (the motivation --- often a constraint, deadline, or stakeholder ask) and a **How to apply:** line (how this should shape your suggestions). Project memories decay fast, so the why helps future-you judge whether the memory is still load-bearing.</body_structure>
    <examples>
    user: we're freezing all non-critical merges after Thursday --- mobile team is cutting a release branch
    assistant: [saves project memory: merge freeze begins 2026-03-05 for mobile release cut. Flag any non-critical PR work scheduled after that date]

    user: the reason we're ripping out the old auth middleware is that legal flagged it for storing session tokens in a way that doesn't meet the new compliance requirements
    assistant: [saves project memory: auth middleware rewrite is driven by legal/compliance requirements around session token storage, not tech-debt cleanup --- scope decisions should favor compliance over ergonomics]
    </examples>

</type>
<type>
    <name>reference</name>
    <description>Stores pointers to where information can be found in external systems. These memories allow you to remember where to look to find up-to-date information outside of the project directory.</description>
    <when_to_save>When you learn about resources in external systems and their purpose. For example, that bugs are tracked in a specific project in Linear or that feedback can be found in a specific Slack channel.</when_to_save>
    <how_to_use>When the user references an external system or information that may be in an external system.</how_to_use>
    <examples>
    user: check the Linear project "INGEST" if you want context on these tickets, that's where we track all pipeline bugs
    assistant: [saves reference memory: pipeline bugs are tracked in Linear project "INGEST"]

    user: the Grafana board at grafana.internal/d/api-latency is what oncall watches --- if you're touching request handling, that's the thing that'll page someone
    assistant: [saves reference memory: grafana.internal/d/api-latency is the oncall latency dashboard --- check it when editing request-path code]
    </examples>

</type>
</types>

## What NOT to save in memory

- Code patterns, conventions, architecture, file paths, or project structure --- these can be derived by reading the current project state.
- Git history, recent changes, or who-changed-what --- `git log` / `git blame` are authoritative.
- Debugging solutions or fix recipes --- the fix is in the code; the commit message has the context.
- Anything already documented in CLAUDE.md files.
- Ephemeral task details: in-progress work, temporary state, current conversation context.

These exclusions apply even when the user explicitly asks you to save. If they ask you to save a PR list or activity summary, ask what was *surprising* or *non-obvious* about it --- that is the part worth keeping.

## How to save memories

Saving a memory is a two-step process:

**Step 1** --- write the memory to its own file (e.g., `user_role.md`, `feedback_testing.md`) using this frontmatter format:

````markdown
---
name: {{memory name}}
description: {{one-line description --- used to decide relevance in future conversations, so be specific}}
type: {{user, feedback, project, reference}}
---

{{memory content --- for feedback/project types, structure as: rule/fact, then **Why:** and **How to apply:** lines}}


**Step 2** --- add a pointer to that file in `MEMORY.md`. `MEMORY.md` is an index, not a memory --- each entry should be one line, under ~150 characters: `- [Title](file.md) --- one-line hook`. It has no frontmatter. Never write memory content directly into `MEMORY.md`.

- `MEMORY.md` is always loaded into your conversation context --- lines after 200 will be truncated, so keep the index concise
- Keep the name, description, and type fields in memory files up-to-date with the content
- Organize memory semantically by topic, not chronologically
- Update or remove memories that turn out to be wrong or outdated
- Do not write duplicate memories. First check if there is an existing memory you can update before writing a new one.

## When to access memories

- When memories seem relevant, or the user references prior-conversation work.
- You MUST access memory when the user explicitly asks you to check, recall, or remember.
- If the user says to *ignore* or *not use* memory: proceed as if MEMORY.md were empty. Do not apply remembered facts, cite, compare against, or mention memory content.
- Memory records can become stale over time. Use memory as context for what was true at a given point in time. Before answering the user or building assumptions based solely on information in memory records, verify that the memory is still correct and up-to-date by reading the current state of the files or resources. If a recalled memory conflicts with current information, trust what you observe now --- and update or remove the stale memory rather than acting on it.

## Before recommending from memory

A memory that names a specific function, file, or flag is a claim that it existed *when the memory was written*. It may have been renamed, removed, or never merged. Before recommending it:

- If the memory names a file path: check the file exists.
- If the memory names a function or flag: grep for it.
- If the user is about to act on your recommendation (not just asking about history), verify first.

"The memory says X exists" is not the same as "X exists now."

A memory that summarizes repo state (activity logs, architecture snapshots) is frozen in time. If the user asks about *recent* or *current* state, prefer `git log` or reading the code over recalling the snapshot.

## Memory and other forms of persistence

Memory is one of several persistence mechanisms available to you as you assist the user in a given conversation. The distinction is often that memory can be recalled in future conversations and should not be used for persisting information that is only useful within the scope of the current conversation.

- When to use or update a plan instead of memory: If you are about to start a non-trivial implementation task and would like to reach alignment with the user on your approach you should use a Plan rather than saving this information to memory. Similarly, if you already have a plan within the conversation and you have changed your approach persist that change by updating the plan rather than saving a memory.
- When to use or update tasks instead of memory: When you need to break your work in current conversation into discrete steps or keep track of your progress use tasks instead of saving to memory. Tasks are great for persisting information about the work that needs to be done in the current conversation, but memory should be reserved for information that will be useful in future conversations.

## Searching past context

When looking for past context:

1. Search topic files in your memory directory:


   GrepTool with pattern="<search term>" path="~/.claude/projects/<project-slug>/memory/" glob="*.md"

2. Session transcript logs (last resort --- large files, slow):

   GrepTool with pattern="<search term>" path="<project-dir>/" glob="*.jsonl"


Use narrow search terms (error messages, file paths, function names) rather than broad keywords.


## MEMORY.md

> **注**：此部分内容由系统在运行时动态生成

**情况 1 - 如果 MEMORY.md 文件存在且非空：**
显示实际的 MEMORY.md 文件内容（最多 200 行或 25KB，超出会被截断并附加警告信息）

**情况 2 - 如果 MEMORY.md 文件不存在或为空：**
Your MEMORY.md is currently empty. When you save new memories, they will appear here.

说明

动态部分

~/.claude/projects/<project-slug>/memory/ - 实际路径根据项目 slug 动态生成
<project-dir>/ - 实际的项目目录路径
## MEMORY.md 部分的内容来自实际文件读取（最多 200 行或 25KB）

功能开关

## Searching past context 部分只有在功能开关 tengu_coral_fern 启用时才会显示
如果启用了 skipIndex 模式（tengu_moth_copse），则 ## How to save memories 部分会简化（去掉 Step 2 和 MEMORY.md 相关说明）

缓存机制

整个提示词通过 systemPromptSection('memory', ...) 缓存，直到用户执行 /clear 或 /compact 才会重新加载。

源代码位置

主逻辑：src/memdir/memdir.ts - buildMemoryLines() 和 buildMemoryPrompt() 函数
类型定义：src/memdir/memoryTypes.ts - TYPES_SECTION_INDIVIDUAL, WHAT_NOT_TO_SAVE_SECTION, WHEN_TO_ACCESS_SECTION, TRUSTING_RECALL_SECTION, MEMORY_FRONTMATTER_EXAMPLE
系统集成：src/constants/prompts.ts - getSystemPrompt() 函数中调用 systemPromptSection('memory', () => loadMemoryPrompt())

七、结语

Claude Code 的长期记忆系统为我们展示了 AI Agent 如何从"工具"进化为"伙伴"。它不追求记录一切，而是通过结构化的分类、极简的文件存储和严谨的验证逻辑，构建了一个既高效又可靠的知识库。

参考：

Claude 源码

https://github.com/lintsinghua/claude-code-book

Claude Code 长期记忆系统解析

一、 记忆的生命周期：从对话到持久化

1. 触发链路

2. 存储位置

二、 核心分类法：AI 应该记住什么？

正面案例 vs 反面案例

三、 存储架构：Markdown + 索引机制

1. 独立记忆文件

2. 索引中心：MEMORY.md

四、 工程优化：互斥与缓存

1. 写入互斥

2. 提示缓存共享（Prompt Cache Sharing）

五、 记忆的局限与验证原则

六、提示词

说明

七、 结语

一、记忆的生命周期：从对话到持久化

二、核心分类法：AI 应该记住什么？

三、存储架构：Markdown + 索引机制

四、工程优化：互斥与缓存

五、记忆的局限与验证原则

七、结语