persistent-memory

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Persistent Memory

持久化内存

Use this skill as the single memory system for this repository.

请将该Skill作为此仓库的唯一内存系统使用。

Commands

命令

Use either command style:

python3 .agents/skills/persistent-memory/scripts/memory.py <command>

.agents/skills/persistent-memory/scripts/pmem <command>

Supported commands:

```
init
```
```
sync
```
(database-only health check)
```
cleanup-legacy
```
```
backfill-embeddings --batch 500
```

prune --source "<label>" [--older-than <days>]

```
search "<query>" --limit 8
```

add "<memory text>" --tags "<comma,tags>" --source "assistant"

```
recent --limit 10
```
```
stats
```

可使用以下任意一种命令格式：

python3 .agents/skills/persistent-memory/scripts/memory.py <command>

.agents/skills/persistent-memory/scripts/pmem <command>

支持的命令：

```
init
```
```
sync
```
（仅数据库健康检查）
```
cleanup-legacy
```
```
backfill-embeddings --batch 500
```

prune --source "<label>" [--older-than <days>]

```
search "<query>" --limit 8
```

add "<memory text>" --tags "<comma,tags>" --source "assistant"

```
recent --limit 10
```
```
stats
```

Required Workflow

必备工作流

Initialize memory in a fresh workspace:

```
pmem init
```

At the start of substantial tasks:

```
pmem sync
```
(database-only health check)

pmem search "<topic keywords>" --limit 8

When user explicitly says
```
remember
```
or when a durable preference/fact is learned:

pmem add "<memory text>" --tags "<tags>" --source "assistant"

Before finalizing memory-sensitive work, verify recall state:

```
pmem stats
```

在新工作区中初始化内存：

```
pmem init
```

在执行重要任务前：

```
pmem sync
```
（仅数据库健康检查）

pmem search "<topic keywords>" --limit 8

当用户明确说出“记住”或获取到持久化偏好/事实时：

pmem add "<memory text>" --tags "<tags>" --source "assistant"

在完成对内存敏感的工作前，验证召回状态：

```
pmem stats
```

One-Time Migration (If Upgrading From Older Setup)

一次性迁移（从旧版本升级时）

Remove legacy imported rows:

```
pmem cleanup-legacy
```

Generate vectors for existing notes:

```
pmem backfill-embeddings
```

删除旧版导入的记录：

```
pmem cleanup-legacy
```

为现有笔记生成向量：

```
pmem backfill-embeddings
```

Storage Rules

存储规则

Store durable preferences, long-lived facts, stable workflows, and repeated constraints.
Do not store noisy one-off transient details unless requested.
Keep entries concise and specific.
Prefer tags that improve retrieval quality (
```
preferences
```
,
```
calendar
```
,
```
comms
```
,
```
product
```
).

存储持久化偏好、长期事实、稳定工作流以及重复出现的约束条件。
除非有要求，否则不要存储杂乱的一次性临时细节。
确保记录简洁且具体。
优先使用能提升检索质量的标签（如
```
preferences
```
、
```
calendar
```
、
```
comms
```
、
```
product
```
）。

Retrieval Rules

检索规则

Use targeted search queries instead of broad terms.
Keep default
```
--limit
```
low unless deeper recall is needed.
```
search
```
automatically reinforces recalled entries by updating
```
hits
```
and
```
last_seen_at
```
.
```
hits
```
are analytics-oriented and not used as a direct ranking boost.
Search uses hybrid retrieval: lexical + semantic.
Semantic search tries
```
sqlite-vec
```
first and auto-falls back to Python cosine if needed.

使用针对性的搜索查询，而非宽泛的术语。
除非需要深度召回，否则保持默认的
```
--limit
```
值较低。
```
search
```
命令会通过更新
```
hits
```
和
```
last_seen_at
```
字段自动强化已召回的记录。
```
hits
```
字段用于分析，不会直接作为排名提升的依据。
搜索采用混合检索方式：词法检索 + 语义检索。
语义检索会优先尝试
```
sqlite-vec
```
，必要时自动回退到Python余弦相似度计算。

Bootstrapping and Recovery

引导与恢复

If
```
.memory/
```
is missing, run
```
pmem init
```
.
```
pmem sync
```
is a lightweight database-only check (no markdown import/export).

If semantic mode degrades, run

pmem stats

to inspect

semantic_backend

and

embedding_coverage

For command examples and quick troubleshooting, read
```
references/usage.md
```
.

如果
```
.memory/
```
目录缺失，请运行
```
pmem init
```
。
```
pmem sync
```
是轻量级的仅数据库检查（不涉及Markdown导入/导出）。
如果语义模式性能下降，请运行
```
pmem stats
```
检查
```
semantic_backend
```
和
```
embedding_coverage
```
。
如需命令示例和快速故障排查，请查阅
```
references/usage.md
```
。