每日总结更新 20260528

This commit is contained in:
xiaoban 2026-05-28 08:00:04 +08:00
parent 2234f3e0ab
commit 68af361960
89 changed files with 99 additions and 17 deletions

View File

@ -126,7 +126,7 @@ System: [...] Feishu[xiaobian] group oc_xxx | 发送者 (ou_xxx) [msg:om_xxx]
| 目录 | 用途 | 说明 |
|------|------|------|
| `memory/` | 短期记忆 | 按天记录工作日志,格式 `YYYY-MM-DD.md` |
| `business_knowledge/` | 业务知识库 | **所有业务知识统一存放于此**包括业务术语、数据表说明、SQL 模板、数据抽取脚本等 |
| `knowledge/` | 业务知识库 | **所有业务知识统一存放于此**包括业务术语、数据表说明、SQL 模板、数据抽取脚本等 |
| `scripts/` | 脚本文件 | 所有 `.py`、`.sh`、`.sql` 等脚本文件必须放在此目录 |
| `output/` | 输出文件 | 所有生成的报表(`.xlsx`、`.csv`)、日志(`.log`)、导出文件等必须放在此目录 |
| `skills/` | 技能定义 | 个人技能目录 |

View File

@ -432,3 +432,11 @@ To https://git.valavala.com/ai_member_only/ai_member_xiaoban
5ace2cc..48953c0 master -> master
[2026-05-26 08:10:02] 工作区备份成功:自动备份 2026-05-26 08:10:01
[2026-05-27 08:10:01] 开始备份工作区...
[master 2234f3e] 自动备份 2026-05-27 08:10:01
2 files changed, 1 insertion(+), 3 deletions(-)
delete mode 100644 tmp_daily_summary.md
remote: . Processing 1 references
remote: Processed 1 references in total
To https://git.valavala.com/ai_member_only/ai_member_xiaoban
48953c0..2234f3e master -> master
[2026-05-27 08:10:02] 工作区备份成功:自动备份 2026-05-27 08:10:01

View File

@ -1,3 +1,5 @@
{"type":"memory.recall.recorded","timestamp":"2026-05-24T02:48:04.923Z","query":"PostgreSQL online database password PG_DB_PASSWORD","resultCount":4,"results":[{"path":"memory/2026-05-24.md","startLine":1,"endLine":30,"score":1},{"path":"memory/2026-05-24.md","startLine":23,"endLine":52,"score":1},{"path":"memory/2026-03-01.md","startLine":1,"endLine":11,"score":1},{"path":"memory/2026-05-24.md","startLine":46,"endLine":61,"score":1}]}
{"type":"memory.recall.recorded","timestamp":"2026-05-25T05:47:41.388Z","query":"账号信息查询 account_id 14157 数据库","resultCount":4,"results":[{"path":"memory/2026-05-24.md","startLine":46,"endLine":71,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-24.md","startLine":23,"endLine":52,"score":1},{"path":"memory/2026-05-24.md","startLine":1,"endLine":30,"score":1}]}
{"type":"memory.recall.recorded","timestamp":"2026-05-26T06:16:58.547Z","query":"账号 account 查询 query 数据库 23765","resultCount":4,"results":[{"path":"memory/2026-05-25.md","startLine":1,"endLine":26,"score":1},{"path":"memory/2026-05-24.md","startLine":46,"endLine":71,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-24.md","startLine":23,"endLine":52,"score":1}]}
{"type":"memory.recall.recorded","timestamp":"2026-05-27T11:20:45.799Z","query":"昆鹏 新skill 发布 技能","resultCount":1,"results":[{"path":"memory/2026-05-25.md","startLine":1,"endLine":26,"score":1}]}
{"type":"memory.recall.recorded","timestamp":"2026-05-27T13:30:03.421Z","query":"vala_knowledgebase 瓦拉星球 游戏剧情 世界观 角色","resultCount":3,"results":[{"path":"memory/2026-05-24.md","startLine":106,"endLine":126,"score":1},{"path":"memory/2026-05-24.md","startLine":85,"endLine":110,"score":1},{"path":"memory/2026-05-24.md","startLine":1,"endLine":30,"score":1}]}

View File

@ -1,6 +1,6 @@
{
"version": 1,
"updatedAt": "2026-05-26T06:16:58.547Z",
"updatedAt": "2026-05-27T13:30:03.421Z",
"entries": {
"memory:memory/2026-05-24.md:1:30": {
"key": "memory:memory/2026-05-24.md:1:30",
@ -9,20 +9,22 @@
"endLine": 30,
"source": "memory",
"snippet": "# 2026-05-24 工作日志 ## 新建技能: studytime-analysis [刘庆逊提出] 创建学习时间分析技能,分析角色完课记录的规律。 ### 技能结构 - `skills/studytime-analysis/SKILL.md` — 技能定义 - `skills/studytime-analysis/scripts/studytime_analysis.py` — Python 分析脚本 ### 分析维度 1. **一周时间分布**排除寒暑假1-2月、7-8月周一~周日各天完课数、时段分布(上午/中午/下午/晚上)、周末是否上课 2. **跨周学习趋势**(包含寒暑假全部数据):总周数、周均完课数、连续性、中断周检测、前后半段趋势对比、突增/骤降检测 3. **完课记录明细表**(全部数据):日期/时间/星期/时段/级别/课程ID ### 数据源 - PostgreSQL Onlinevala 库) - 核心表:`user_chapter_play_record_0~7`8张分表无 `bi_` 前缀) - 筛选:`play_status = 1` - 注意:表在 PostgreSQL 而非 MySQL表名无 `bi_` 前缀 ### 寒暑假规则 - 一周分布分析时排除 1-2 月(寒假)和 7-8 月(暑假)—— 因为寒暑假作息与平时差异大,混在一起会干扰时段分析 - 跨周趋势和明细表包含全部数据(含寒暑假) - 报告中区分标注数据范围 ### 触发方式 用户说「学习时间分析 [角色ID]」即可触发 ### 已测试角色",
"recallCount": 2,
"recallCount": 3,
"dailyCount": 0,
"groundedCount": 0,
"totalScore": 2,
"totalScore": 3,
"maxScore": 1,
"firstRecalledAt": "2026-05-24T02:48:04.923Z",
"lastRecalledAt": "2026-05-25T05:47:41.388Z",
"lastRecalledAt": "2026-05-27T13:30:03.421Z",
"queryHashes": [
"c2d15f7574fb",
"9aff8ec9594a"
"9aff8ec9594a",
"71463fe40be2"
],
"recallDays": [
"2026-05-24",
"2026-05-25"
"2026-05-25",
"2026-05-27"
],
"conceptTags": [
"studytime-analysis",
@ -172,20 +174,22 @@
"endLine": 110,
"source": "memory",
"snippet": "- `skills/studytime-analysis/scripts/studytime_analysis.py` — 重写 `fetch_chapter_info_map()`,新增全局 unit_index 计算HTML 模板更新为 Level/Unit/Lesson 三列 - 已为角色 32009zyl重新生成 HTML 并发送 - 已同步 Git + SkillHub ## 新建技能: studycourse-analysis (2026-05-24) [刘庆逊提出] 创建角色上课情况分析技能,从四维度分析角色学习数据。 ### 技能结构 - `skills/studycourse-analysis/SKILL.md` — 技能定义 - `skills/studycourse-analysis/scripts/studycourse_analysis.py` — Python 分析脚本 ### 四步分析 1. **基础信息**:角色姓名/年龄/账号ID/手机号后4位/注册时间/购买渠道/设备/首末次完课 2. **完课耗时**:平均值/中位数、异常检测(<10min / >20min)、前后半段趋势 3. **中互动正确率**Perfect/Good/Oops/Pass/Failed 占比和趋势 4. **知识巩固**:完成率、正确率得分分布 ### 数据源 | 类型 | 库 | 表 | 用途 | |------|-----|-----|------| | MySQL vala_user | vala_app_character | 角色信息、pu",
"recallCount": 2,
"recallCount": 3,
"dailyCount": 0,
"groundedCount": 0,
"totalScore": 2,
"totalScore": 3,
"maxScore": 1,
"firstRecalledAt": "2026-05-25T05:47:41.388Z",
"lastRecalledAt": "2026-05-26T06:16:58.547Z",
"lastRecalledAt": "2026-05-27T13:30:03.421Z",
"queryHashes": [
"9aff8ec9594a",
"566b5958861e"
"566b5958861e",
"71463fe40be2"
],
"recallDays": [
"2026-05-25",
"2026-05-26"
"2026-05-26",
"2026-05-27"
],
"conceptTags": [
"fetch-chapter-info-map",
@ -205,18 +209,20 @@
"endLine": 26,
"source": "memory",
"snippet": "# 2026-05-25 工作日志 ## user-info 技能重写 [刘庆逊提出] 修复 `user-info` 技能,使其匹配线上实际数据库结构。 ### 问题 旧脚本引用的表(`bi_vala_app_account`、`account_login`、`account_detail_info`、`bi_vala_order`、`bi_level_unit_lesson`)在线上数据库均不存在。 ### 修复内容 - **scripts/query_user_info.py** 完整重写: - 表名改为实际线上表:`vala_user.vala_app_account`、`vala_user.vala_app_character`、`vala_order.vala_seasonal_ticket`、PG `user_chapter_play_record_0~7` - 手机号查询通过 `tel LIKE '前缀%后缀'` 脱敏匹配 - Chapter → Level/Unit/Lesson 映射复用 studytime-analysis 的 `fetch_chapter_info_map()` 逻辑 - 订单数据改用 `vala_seasonal_ticket`(赛季通票),因线上无标准订单表 - 设备/地域信息标注为暂不可用(线上无对应表) - PG 时区处理:`created_at` 为 tz-aware统一转 naive 比较 - **SKILL.md** 更新至 v2.0.0,补充数据覆盖说明 - **references/dat",
"recallCount": 1,
"recallCount": 2,
"dailyCount": 0,
"groundedCount": 0,
"totalScore": 1,
"totalScore": 2,
"maxScore": 1,
"firstRecalledAt": "2026-05-26T06:16:58.547Z",
"lastRecalledAt": "2026-05-26T06:16:58.547Z",
"lastRecalledAt": "2026-05-27T11:20:45.799Z",
"queryHashes": [
"566b5958861e"
"566b5958861e",
"ef1f12a9b060"
],
"recallDays": [
"2026-05-26"
"2026-05-26",
"2026-05-27"
],
"conceptTags": [
"user-info",
@ -228,6 +234,37 @@
"scripts/query-user-info.py",
"vala-user.vala-app-account"
]
},
"memory:memory/2026-05-24.md:106:126": {
"key": "memory:memory/2026-05-24.md:106:126",
"path": "memory/2026-05-24.md",
"startLine": 106,
"endLine": 126,
"source": "memory",
"snippet": "| MySQL vala_user | vala_app_character | 角色信息、purchase_season_package | | MySQL vala_user | vala_app_account | 下载渠道、手机号、注册时间 | | MySQL vala | vala_game_chapter + season_package | 章节映射 | | PostgreSQL vala | user_course_detail | 课程激活/到期时间 | | PostgreSQL vala | user_login_app_info | 设备信息 | | PostgreSQL vala | user_chapter_play_record_0~7 | 完课记录(play_status=1) | | PostgreSQL vala | user_component_play_record_0~7 | 中互动记录(play_result) | | PostgreSQL vala | user_chapter_settlement_data_0~7 | 巩固数据(settlement_data JSON) | ### 关键发现 - **设备信息**来自 `user_login_app_info`device_name/model/type/os_info/city - **购买渠道**来自 `vala_app_account.download_channel` + `key_from` - **巩固判断**`settlement_data.practiceS",
"recallCount": 1,
"dailyCount": 0,
"groundedCount": 0,
"totalScore": 1,
"maxScore": 1,
"firstRecalledAt": "2026-05-27T13:30:03.421Z",
"lastRecalledAt": "2026-05-27T13:30:03.421Z",
"queryHashes": [
"71463fe40be2"
],
"recallDays": [
"2026-05-27"
],
"conceptTags": [
"vala-user",
"vala-app-character",
"purchase-season-package",
"vala-app-account",
"vala-game-chapter",
"season-package",
"user-course-detail",
"课程激活/到期时间"
]
}
}
}

32
memory/2026-05-27.md Normal file
View File

@ -0,0 +1,32 @@
# 2026-05-27 工作日志
## 小红书渠道财务预测模型(陈逸鸫)
- [陈逸鸫] 需要搭建小红书渠道 2026年6-12月的财务预测模型
- 关键参数:客单价 ¥2,500、退费率 3%、销售佣金 10%、月进量基准 1750 条、转化率基准 10%、CPL 275 元、无固定成本
- K12 季节性系数自行设定7月/8月/12月为大月
- 用 sonnet 模型搭框架,后续用 opus 或 deepseek 做深度分析
- **结论**当前参数下单元经济亏损CAC ¥2,750 > 单客净收入 ¥2,182.5),盈亏平衡需 CVR≥12.6% 或 ASP≥¥3,150 或 CPL≤¥218
- **待确认**:用户需提供可调整的参数方向
## 瓦拉星球漫画化方案(陈逸鸫)
- [陈逸鸫] 希望将瓦拉星球游戏剧情做成日漫风连载漫画,发布到小红书,用于拓展品牌人群资产,引导英语课程试听
- 已探索飞书 VALA Academy wiki spacespace_id: `7487521512283799556`),获取了完整游戏世界观:
- 四大势力地球地下城、艾斯环空间站、万得港海盗巢、Vala 新大陆
- 核心角色Fio15 岁女海盗王、Nook 鳄鱼思维芯片联合发明人被背叛、Lauren主角母亲失踪/牺牲、Pemberton反派/校长)、主角(开拓者后代)
- 经过多次迭代优化开场方案:
- ❌ 第1版芯片悬念开场信息过载陌生信息太多
- ❌ 第2版《妈妈失踪的那天》慢铺垫信息太慢冲突不够
- ✅ 第3版Nook 背叛发布会、Fio 通缉令、Pemberton 杀母档案 三个冷开场
- ✅ **最终版**《万得港》6格——妈妈失踪十年 → 收到 F 的加密邮件线索 → 冲向万得港 → 新消息揭示"她不是偶然在那里的"
- 已规划 12 话三幕结构HOOK → BUILD → CONVERT转化节点设计为主角收到妈妈英文语音 → Nook 教英语 → 扫码听课解锁第 4 个词
- 已生成第 1 话封面图日漫画风2K 分辨率3:4 竖版)
- **待定**:下一步是生成内容页 / 角色定妆照 / 小红书发布文案
## 工具/技能发现
- 确认昆鹏发布的 `multi-model-assistant``gpt-image-2-generator` 两个技能可用
- `multi-model-assistant` 支持调用多个模型sonnet/opus/deepseek/gpt5.4 等),位于 `/root/.openclaw/skills/multi-model-assistant/`
- `gpt-image-2-generator` 通过 FunCloud API 异步生成图片,位于 `/root/.openclaw/skills/gpt-image-2-generator/`
- 语音转写使用 DashScope qwen-audio-turbo APIKey 为 `DASHSCOPE_API_KEY`

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.1 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.9 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.5 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.5 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.5 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.2 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.0 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.7 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.2 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 5.2 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.9 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.6 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.5 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 3.4 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.8 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.8 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.4 MiB

Binary file not shown.

After

Width:  |  Height:  |  Size: 4.1 MiB

3
tmp_daily_summary.md Normal file
View File

@ -0,0 +1,3 @@
=== 每日总结 20260528 ===
## 昨日关键进展
- 经过多次迭代优化开场方案: