memory-init

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Purpose

目标

Initialize the Memory Hub for a project: create

.memory/

skeleton, scan the project, and populate initial knowledge files and catalog.

为项目初始化Memory Hub：创建

.memory/

目录骨架，扫描项目并填充初始知识库文件与目录。

Input

输入

No arguments required. Run from the project root.

无需传入参数。请在项目根目录执行。

Required Flow

必要流程

Step 1: Create skeleton

步骤1：创建目录骨架

bash

memory-hub init

ALREADY_INITIALIZED

error, stop and inform the user.

bash

memory-hub init

如果出现

ALREADY_INITIALIZED

错误，停止操作并通知用户。

Step 2: Scan project and generate tech-stack knowledge

步骤2：扫描项目并生成技术栈知识库

Read the project root file listing

Read package manager files (

package.json

pyproject.toml

Cargo.toml

go.mod

pom.xml

, etc.)

Read entry files (
```
main.*
```
,
```
index.*
```
,
```
app.*
```
,
```
server.*
```
, etc.)

Read config files (

.env.example

tsconfig.json

webpack.config.*

, etc.)

Generate

tech-stack.md

content and write:

bash

memory-hub write architect tech-stack.md \
  --topic tech-stack --summary "技术栈、关键依赖、使用方式与限制" --mode overwrite <<'EOF'
<generated markdown>
EOF

读取包管理文件（

package.json

、

pyproject.toml

、

Cargo.toml

、

go.mod

、

pom.xml

等）

读取入口文件（
```
main.*
```
、
```
index.*
```
、
```
app.*
```
、
```
server.*
```
等）

读取配置文件（

.env.example

、

tsconfig.json

、

webpack.config.*

等）

生成

tech-stack.md

内容并写入：

bash

memory-hub write architect tech-stack.md \
  --topic tech-stack --summary "技术栈、关键依赖、使用方式与限制" --mode overwrite <<'EOF'
<generated markdown>
EOF

Step 3: Generate code conventions knowledge

步骤3：生成代码约定知识库

Based on project structure from Step 2, generate

conventions.md

bash

memory-hub write dev conventions.md \
  --topic conventions --summary "目录命名规则、模块组织方式、代码约定" --mode overwrite <<'EOF'
<generated markdown>
EOF

基于步骤2得到的项目结构，生成

conventions.md

：

bash

memory-hub write dev conventions.md \
  --topic conventions --summary "目录命名规则、模块组织方式、代码约定" --mode overwrite <<'EOF'
<generated markdown>
EOF

Step 4: Scan project modules and generate Catalog

步骤4：扫描项目模块并生成目录

Analyze directory structure, identify functional domains and key files, construct JSON:

bash

memory-hub catalog-update <<'EOF'
{"modules": [{"name": "...", "summary": "...", "files": [{"path": "...", "description": "..."}]}]}
EOF

分析目录结构，识别功能领域与关键文件，构建JSON：

bash

memory-hub catalog-update <<'EOF'
{"modules": [{"name": "...", "summary": "...", "files": [{"path": "...", "description": "..."}]}]}
EOF

Step 5: Quality gate

步骤5：质量校验

List
```
unknowns
```
— files/directories that don't clearly belong to any functional domain
Check
```
catalog.repair
```
output (auto-triggered by catalog-update):
- ```
ai_actions
```
  non-empty → execute self-healing, then run
```
memory-hub catalog-repair
```
  again to confirm cleared
- ```
manual_actions
```
  non-empty → report to user

列出
```
unknowns
```
— 无法明确归属到任何功能领域的文件/目录
检查
```
catalog.repair
```
的输出（由catalog-update自动触发）：
- ```
ai_actions
```
  非空 → 执行自我修复，然后再次运行
```
memory-hub catalog-repair
```
  确认已清理
- ```
manual_actions
```
  非空 → 向用户报告

Step 6: Output summary

步骤6：输出摘要

Report: files created, modules identified, unknowns requiring user confirmation.

报告内容：已创建的文件、已识别的模块、需要用户确认的未知项。

Output

输出

JSON envelope from each command. Final summary to user.

每个命令返回JSON格式结果。最终向用户展示摘要信息。

Error Handling

错误处理

```
ALREADY_INITIALIZED
```
→ inform user, do not proceed
Any command failure → stop and report the error

```
ALREADY_INITIALIZED
```
→ 通知用户，不再继续执行
任何命令执行失败 → 停止操作并报告错误