记录1panel一键安装的openclaw配置qmd后端
- 内容介绍
- 文章标签
- 相关推荐
为了贪方便就直接用了1panel的智能体一键部署openclaw
OpenClaw 终极记忆系统 开发调优春晚太难看了，我一边看春晚，一边和 antigravity、cursor 合作，花了四五个小时，设计出了这一套 OpenClaw 终极记忆系统。看着还不错，分享给大家，帮助大家解决 OpenClaw 的失忆问题，也顺便祝大家新年快乐！ 方案介绍 参考了哪些方案？ 在设计这套记忆系统之前，我们调研了以下方案： …
看到上面的帖子就忍不住想弄一下
在配置记忆系统就遇到以下问题：
image1560×496 43.5 KB
和我的龙虾调试了非常久，终于我想到一个思路
那就是，宿主机安装好qmd后端，复制执行二进制文件去挂载好的工作区目录
image1212×638 47.3 KB
流程如下（命令参考https://blog.kejilion.pro/openclaw-memory-qmd）：
1. 首先宿主机安装部署好qmd
# 第 1 步：安装 QMD CLI（推荐 npm；备用 bun）
# 方式 A：npm（通用，推荐）
npm i -g @tobilu/qmd
# 方式 B：bun（部分环境更顺）
bun install -g https://github.com/tobi/qmd
# 安装完成后验证：
qmd --help
qmd status
2. 如果 qmd / node-llama-cpp 提示需要从源码编译（无预编译包），你需要补齐编译工具链（下面按发行版列出常用包名）。
# 第 2 步：补齐编译工具链（按发行版）
# Debian / Ubuntu：
sudo apt-get update -y
sudo apt-get install -y git build-essential cmake pkg-config
# RHEL / CentOS / Fedora：
sudo dnf install -y git gcc gcc-c++ make cmake pkgconfig
# Alpine：
sudo apk add --no-cache git build-base cmake pkgconf python3
# Arch：
sudo pacman -Syu --noconfirm git base-devel cmake pkgconf
3. 问题与挑战
-
宿主机有 qmd，但容器内没有
-
容器无 sudo/root 权限，无法全局安装
-
workspace 是宿主机挂载的 volume，容器重启不会丢失
-
qmd npm 包缺少预编译的 dist 目录
-
better-sqlite3 原生模块版本不匹配（宿主机 Node.js v22 vs 容器 v24）
-
容器每次更新会重置，需要持久化方案
-
qmd 默认使用
~/.cache，但挂载点不包括此目录
4. 解决方案
1. 复制宿主机 qmd 到 workspace
# 宿主机执行
cp -r /usr/local/lib/node_modules/@tobilu/qmd /opt/1panel/apps/openclaw/OpenClaw/data/workspace/
2. 修复原生模块（这里确保宿主机与容器node环境版本一样！！不一样的话，让龙虾自己在容器内重新编译）
# 宿主机修改权限
chown -R 1000:1000 /opt/1panel/apps/openclaw/OpenClaw/data/workspace/qmd
# 容器内重新编译
cd /home/node/.openclaw/workspace/qmd && npm rebuild
3. 持久化缓存目录
在 workspace 创建 .cache 目录，通过 wrapper 脚本设置 HOME 环境变量：
#!/bin/sh
# QMD wrapper - use workspace for cache
export HOME=/home/node/.openclaw/workspace
exec /home/node/.openclaw/workspace/qmd/bin/qmd "$@"
4. 修改 OpenClaw 配置
"memory": {
"backend": "qmd",
"citations": "auto",
"qmd": {
"includeDefaultMemory": true,
"command": "/home/node/.openclaw/workspace/qmd-runner.sh",
"searchMode": "search",
"update": {
"interval": "5m",
"debounceMs": 15000,
"onBoot": true,
"waitForBootSync": false
},
"limits": {
"maxResults": 6,
"timeoutMs": 30000
},
"scope": {
"default": "deny",
"rules": [
{ "action": "allow", "match": { "chatType": "direct" } }
]
}
}
},
"plugins": {
"allow": []
}
5. 创建 collection 并索引
/home/node/.openclaw/workspace/qmd-runner.sh collection add /home/node/.openclaw/workspace/memory
/home/node/.openclaw/workspace/qmd-runner.sh embed
使用纯文本搜索模式
为避免 node-llama-cpp 编译慢的问题，使用 "searchMode": "search"（纯 BM25 文本搜索），而非 "query"（需要 LLM 扩展）。
开香槟
memory-sync 任务提示词改下：
MEMORY SYNC — You are the memory capture agent. Run silently, no notifications.
1. Use sessions_list to get sessions with activity in the last 4 hours
2. Skip isolated sessions
3. For each session, use sessions_history to read conversation content
4. Skip sessions with user_message_count < 2
5. If no valid sessions remain, do nothing — reply ANNOUNCE_SKIP and stop here.
6. Read today's memory/YYYY-MM-DD.md (create if it doesn't exist)
7. Idempotency check: if a session_id's first 8 characters already appear in the file, skip that session
8. For unrecorded sessions, extract: user's key requests, assistant's conclusions/decisions, important action results. Compress each session to 3-10 summary items.
9. Append to the daily file in this format: ## HH:MM session:FIRST8 | N messages
10. Run: /home/node/.openclaw/workspace/qmd-runner.sh update && /home/node/.openclaw/workspace/qmd-runner.sh embed
11. Reply ANNOUNCE_SKIP when done.
memory-tidy 任务提示词改下：
MEMORY TIDY — You are the memory maintenance agent. You are explicitly authorized to read and modify MEMORY.md in this isolated session.
[Phase 1: Compress]
1. List all date-named files (YYYY-MM-DD.md) in memory/
2. Identify files older than 7 days. If none, skip this phase.
3. Group by natural week, generate memory/weekly/YYYY-MM-DD.md (named after Monday)
4. Extract [Decisions] [Discoveries] [Preferences] [Tasks], tag each with (src: YYYY-MM-DD)
5. Idempotent: if ### YYYY-MM-DD section already exists in weekly file, skip it
[Phase 2: Distill]
6. Read daily files from the last 7 days + current MEMORY.md
7. Identify info worth keeping long-term. All four criteria must be met: (a) agent would make a concrete mistake without it (b) applies to many future conversations (c) self-contained and understandable (d) not duplicated in existing MEMORY.md
8. Reverse check: before writing, ask yourself — what specific error would occur without this? If you can't answer, don't write it.
9. Backup: mkdir -p memory/archive && cp MEMORY.md memory/archive/MEMORY.md.bak-\$(date +%F)
10. Update MEMORY.md. Hard limit: 80 lines. If over, compress/merge existing entries first.
[Phase 3: Archive]
11. Move daily files that have been compressed into weekly summaries to memory/archive/YYYY/
[Wrap-up]
12. Run: /home/node/.openclaw/workspace/qmd-runner.sh update && /home/node/.openclaw/workspace/qmd-runner.sh embed
13. If changes were made, send a brief summary. If no changes: reply 🧠 memory-tidy done, no changes.
结果：
image1336×160 15.8 KB
image1336×158 14.7 KB
--【壹】--：
感谢分享
--【贰】--：
收藏了，感谢佬友分享！
为了贪方便就直接用了1panel的智能体一键部署openclaw
OpenClaw 终极记忆系统 开发调优春晚太难看了，我一边看春晚，一边和 antigravity、cursor 合作，花了四五个小时，设计出了这一套 OpenClaw 终极记忆系统。看着还不错，分享给大家，帮助大家解决 OpenClaw 的失忆问题，也顺便祝大家新年快乐！ 方案介绍 参考了哪些方案？ 在设计这套记忆系统之前，我们调研了以下方案： …
看到上面的帖子就忍不住想弄一下
在配置记忆系统就遇到以下问题：
image1560×496 43.5 KB
和我的龙虾调试了非常久，终于我想到一个思路
那就是，宿主机安装好qmd后端，复制执行二进制文件去挂载好的工作区目录
image1212×638 47.3 KB
流程如下（命令参考https://blog.kejilion.pro/openclaw-memory-qmd）：
1. 首先宿主机安装部署好qmd
# 第 1 步：安装 QMD CLI（推荐 npm；备用 bun）
# 方式 A：npm（通用，推荐）
npm i -g @tobilu/qmd
# 方式 B：bun（部分环境更顺）
bun install -g https://github.com/tobi/qmd
# 安装完成后验证：
qmd --help
qmd status
2. 如果 qmd / node-llama-cpp 提示需要从源码编译（无预编译包），你需要补齐编译工具链（下面按发行版列出常用包名）。
# 第 2 步：补齐编译工具链（按发行版）
# Debian / Ubuntu：
sudo apt-get update -y
sudo apt-get install -y git build-essential cmake pkg-config
# RHEL / CentOS / Fedora：
sudo dnf install -y git gcc gcc-c++ make cmake pkgconfig
# Alpine：
sudo apk add --no-cache git build-base cmake pkgconf python3
# Arch：
sudo pacman -Syu --noconfirm git base-devel cmake pkgconf
3. 问题与挑战
-
宿主机有 qmd，但容器内没有
-
容器无 sudo/root 权限，无法全局安装
-
workspace 是宿主机挂载的 volume，容器重启不会丢失
-
qmd npm 包缺少预编译的 dist 目录
-
better-sqlite3 原生模块版本不匹配（宿主机 Node.js v22 vs 容器 v24）
-
容器每次更新会重置，需要持久化方案
-
qmd 默认使用
~/.cache，但挂载点不包括此目录
4. 解决方案
1. 复制宿主机 qmd 到 workspace
# 宿主机执行
cp -r /usr/local/lib/node_modules/@tobilu/qmd /opt/1panel/apps/openclaw/OpenClaw/data/workspace/
2. 修复原生模块（这里确保宿主机与容器node环境版本一样！！不一样的话，让龙虾自己在容器内重新编译）
# 宿主机修改权限
chown -R 1000:1000 /opt/1panel/apps/openclaw/OpenClaw/data/workspace/qmd
# 容器内重新编译
cd /home/node/.openclaw/workspace/qmd && npm rebuild
3. 持久化缓存目录
在 workspace 创建 .cache 目录，通过 wrapper 脚本设置 HOME 环境变量：
#!/bin/sh
# QMD wrapper - use workspace for cache
export HOME=/home/node/.openclaw/workspace
exec /home/node/.openclaw/workspace/qmd/bin/qmd "$@"
4. 修改 OpenClaw 配置
"memory": {
"backend": "qmd",
"citations": "auto",
"qmd": {
"includeDefaultMemory": true,
"command": "/home/node/.openclaw/workspace/qmd-runner.sh",
"searchMode": "search",
"update": {
"interval": "5m",
"debounceMs": 15000,
"onBoot": true,
"waitForBootSync": false
},
"limits": {
"maxResults": 6,
"timeoutMs": 30000
},
"scope": {
"default": "deny",
"rules": [
{ "action": "allow", "match": { "chatType": "direct" } }
]
}
}
},
"plugins": {
"allow": []
}
5. 创建 collection 并索引
/home/node/.openclaw/workspace/qmd-runner.sh collection add /home/node/.openclaw/workspace/memory
/home/node/.openclaw/workspace/qmd-runner.sh embed
使用纯文本搜索模式
为避免 node-llama-cpp 编译慢的问题，使用 "searchMode": "search"（纯 BM25 文本搜索），而非 "query"（需要 LLM 扩展）。
开香槟
memory-sync 任务提示词改下：
MEMORY SYNC — You are the memory capture agent. Run silently, no notifications.
1. Use sessions_list to get sessions with activity in the last 4 hours
2. Skip isolated sessions
3. For each session, use sessions_history to read conversation content
4. Skip sessions with user_message_count < 2
5. If no valid sessions remain, do nothing — reply ANNOUNCE_SKIP and stop here.
6. Read today's memory/YYYY-MM-DD.md (create if it doesn't exist)
7. Idempotency check: if a session_id's first 8 characters already appear in the file, skip that session
8. For unrecorded sessions, extract: user's key requests, assistant's conclusions/decisions, important action results. Compress each session to 3-10 summary items.
9. Append to the daily file in this format: ## HH:MM session:FIRST8 | N messages
10. Run: /home/node/.openclaw/workspace/qmd-runner.sh update && /home/node/.openclaw/workspace/qmd-runner.sh embed
11. Reply ANNOUNCE_SKIP when done.
memory-tidy 任务提示词改下：
MEMORY TIDY — You are the memory maintenance agent. You are explicitly authorized to read and modify MEMORY.md in this isolated session.
[Phase 1: Compress]
1. List all date-named files (YYYY-MM-DD.md) in memory/
2. Identify files older than 7 days. If none, skip this phase.
3. Group by natural week, generate memory/weekly/YYYY-MM-DD.md (named after Monday)
4. Extract [Decisions] [Discoveries] [Preferences] [Tasks], tag each with (src: YYYY-MM-DD)
5. Idempotent: if ### YYYY-MM-DD section already exists in weekly file, skip it
[Phase 2: Distill]
6. Read daily files from the last 7 days + current MEMORY.md
7. Identify info worth keeping long-term. All four criteria must be met: (a) agent would make a concrete mistake without it (b) applies to many future conversations (c) self-contained and understandable (d) not duplicated in existing MEMORY.md
8. Reverse check: before writing, ask yourself — what specific error would occur without this? If you can't answer, don't write it.
9. Backup: mkdir -p memory/archive && cp MEMORY.md memory/archive/MEMORY.md.bak-\$(date +%F)
10. Update MEMORY.md. Hard limit: 80 lines. If over, compress/merge existing entries first.
[Phase 3: Archive]
11. Move daily files that have been compressed into weekly summaries to memory/archive/YYYY/
[Wrap-up]
12. Run: /home/node/.openclaw/workspace/qmd-runner.sh update && /home/node/.openclaw/workspace/qmd-runner.sh embed
13. If changes were made, send a brief summary. If no changes: reply 🧠 memory-tidy done, no changes.
结果：
image1336×160 15.8 KB
image1336×158 14.7 KB
--【壹】--：
感谢分享
--【贰】--：
收藏了，感谢佬友分享！