morning-briefing 晨间简报自动化工作流 | 个人生产力与信息聚合工具

morning-briefing by krishagel/geoffrey

129 周安装量

4 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/krishagel/geoffrey --skill morning-briefing

信息聚合自动化生产力

🇨🇳中文介绍

晨间简报工作流

概述

生成每日晨间简报，并通过三种方式交付：

终端：立即显示摘要
Obsidian：完整简报保存至每日笔记
电子邮件：简报文本 + 音频播客附件

位置

华盛顿州吉格港 - 用于天气查询

第一阶段：收集数据

执行以下数据收集步骤：

1.1 日历（今日事件）

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun calendar/list_events.js psd --today

返回包含今日事件的 JSON，包括：

事件摘要、地点、开始/结束时间
参与者和回复状态
Hangout/Meet 链接

账户：使用 psd 作为工作日历

1.2 OmniFocus 任务（到期和已标记）

osascript -l JavaScript /Users/hagelk/non-ic-code/geoffrey/skills/morning-briefing/scripts/get_due_flagged.js

返回以下任务：

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

1.3 近期邮件（过去 24 小时）

# 获取 Gmail 查询格式的昨天日期
YESTERDAY=$(date -v-1d +%Y/%m/%d)
cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun gmail/list_messages.js psd --query "in:inbox after:$YESTERDAY" --max 15

返回近期收件箱消息，包含：

发件人、主题、日期、摘要
已读或未读状态
用于上下文的线程 ID
标签数组

过滤：仅显示仍在收件箱中的邮件（尚未被标记/处理）。

带有自定义标签（Label_XXXXX）的邮件已被分类，应排除
例外：包含 SaneCC 或 SaneLater 标签的邮件（值得了解）
系统标签（UNREAD、INBOX、CATEGORY_*、IMPORTANT）不计为“已处理”

理念：过去 24 小时内仍在收件箱中的任何邮件都需要关注，无论是否已读。

账户：使用 psd 作为工作邮箱

1.4 开放的 Freshservice 工单

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/list_tickets.js '{"workspace_id": 2, "filter": "new_and_my_open"}'

返回分配给用户或在技术工作区中由用户创建的开放工单。

1.5 待处理的审批

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_approvals.js requested

返回等待审批的服务请求。

Gig Harbor WA weather today forecast

当前状况
最高/最低温度
降水概率
任何警报

1.7 教育科技新闻（附摘要）

K-12 education technology news past 24 hours January 2026

对于找到的每篇文章（3-5 篇）：

使用 WebFetch 阅读全文
提取 2-3 句摘要，涵盖： * 发生了什么 / 有什么新内容 * 为什么这对 K-12 教育很重要 * 关键要点或行动项

新鲜度检查：在包含前验证文章日期。仅包含过去 24 小时内发布的文章。即使出现在结果中，也跳过较旧的文章。

教育科技产品发布和更新
学校技术政策变化
学校网络安全
数字学习趋势

1.8 人工智能新闻（附摘要）

artificial intelligence news past 24 hours January 2026 latest

对于找到的每篇文章（3-5 篇）：

使用 WebFetch 阅读全文
提取 2-3 句摘要，涵盖： * 是什么发展 * 行业影响 * 与教育/工作的相关性

新鲜度检查：在包含前验证文章日期。仅包含过去 24 小时内发布的文章。即使出现在结果中，也跳过较旧的文章。

主要 AI 模型发布和能力
AI 政策和法规
企业 AI 采用
AI 研究突破

1.9 K-12 领导力新闻（附摘要）

K-12 school leadership superintendent news past 24 hours January 2026

对于找到的每篇文章（2-3 篇）：

使用 WebFetch 阅读全文
提取 2-3 句摘要，涵盖： * 政策或领导力发展 * 对学区/学校的影响 * 与 CIO/技术领导力的相关性

新鲜度检查：在包含前验证文章日期。仅包含过去 24 小时内发布的文章。即使出现在结果中，也跳过较旧的文章。

州和联邦教育政策
学监和董事会新闻
学校资金和预算
劳动力和人员配置趋势

1.9b 学校安全与安保新闻（附摘要）

K-12 school safety security news past 24 hours January 2026

对于找到的每篇文章（3 篇）：

使用 WebFetch 阅读全文
提取 2-3 句摘要，涵盖： * 发生了什么 / 有什么新内容 * 对学校安全实践的影响 * 学区的关键要点

新鲜度检查：在包含前验证文章日期。仅包含过去 24 小时内发布的文章。即使出现在结果中，也跳过较旧的文章。

学校安全事件和响应
安全政策变化
应急准备
心理健康和威胁评估
物理安全技术

1.10 技术团队日终消息（上一个工作日）- 详细版

从技术员工空间获取日终签到消息：

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun chat/get_eod_messages.js psd spaces/AAAAxOtpv10 last-business-day

关键 - 严禁虚构：

仅使用脚本输出 sender 字段中出现的姓名
如果脚本未返回任何消息，则说“未找到日终消息”
如果脚本返回的消息中发件人是 ID 而非姓名，则显示该 ID（例如 "users/12345..."）
严禁虚构姓名、地点或成就
采用复制粘贴方式：将脚本输出视为唯一真实来源

在提取团队数据之前：

检查脚本的原始 JSON 输出
列出返回的确切 sender 值
仅使用这些姓名 - 不要意译，不要“改进”

关键：从每位团队成员的日终消息中提取完整细节。查找包含 "Today:" 前缀的消息 - 这些是详细的日终摘要。

对于每位发布了日终摘要的团队成员：

姓名：发布者（完全使用 sender 字段中的姓名）
地点：他们在哪里工作（居家办公、DCRC、学校站点）
关键成就：完成的具体任务（不仅仅是“处理工单”）
重要事项：解决的有趣问题、参与的项目
问题/阻碍：提到的任何问题

从原始消息中提取的示例：

Brad White:
- Location: TSD (Tech Services)
- Accomplished: Packaged Cinema 4D plugin for deployment, fixed OAuth blocking for Maxon App sign-in, created SwiftDialog notification for plugin installs, used Claude Code for first time to create Installomator label for Godot game engine
- Notable: Working on Unity deployment troubleshooting, burning comp time leaving early

注意：如果今天是周一，“上一个工作日” = 周五（如果周一是假日，则为周四）。

1.10b 安全与安保团队日终消息（上一个工作日）

从安全与安保员工空间获取日终签到消息：

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun chat/get_eod_messages.js psd spaces/AAAAFpQaAnA last-business-day

空间：PSD 安全与安保团队

关键 - 严禁虚构：

仅使用脚本输出 sender 字段中出现的姓名
如果脚本未返回任何消息，则说“未找到日终消息”
如果脚本返回的消息中发件人是 ID（例如 "users/12345..."）而非姓名：
- 显示为：未知 (users/123456...) - [来自消息内容的地点]
- 添加备注：“姓名未在映射中 - 更新 chat_user_mapping.json”
严禁虚构姓名、地点或成就
采用复制粘贴方式：将脚本输出视为唯一真实来源

在提取团队数据之前：

检查脚本的原始 JSON 输出
列出返回的确切 sender 值
仅使用这些姓名 - 不要意译，不要“改进”

使用与技术团队相同的格式提取 - 姓名、地点、成就、问题。

注意：如果今天是周一，“上一个工作日” = 周五（如果周一是假日，则为周四）。

1.11 团队完成的工单（上一个工作日）

获取技术团队在上一个工作日关闭的工单：

# 首先获取上一个工作日
LAST_BIZ_DAY=$(bun /Users/hagelk/non-ic-code/geoffrey/skills/morning-briefing/scripts/get_last_business_day.js | jq -r '.date')

# 然后获取该日期的每日摘要
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "$LAST_BIZ_DAY"

关键 - 严禁虚构：

仅使用脚本输出 byAgent 部分中出现的代理姓名
完全使用脚本返回的工单数量
严禁虚构姓名或工单数量
如果代理姓名显示为 ID 或电子邮件，则按原样显示

在报告工单统计信息之前：

检查 get_daily_summary.js 的原始 JSON 输出
列出 byAgent 字段中的确切代理姓名
仅使用这些姓名和数量

关闭的工单总数
按代理细分
按类别细分（密码重置、Chromebook 等）
自动解决与代理解决

工作区：技术（workspace_id: 2）

1.12 工单趋势分析

将上一个工作日与前几天进行比较以分析趋势：

# 获取最近 5 个工作日的摘要以进行趋势分析
# get_daily_summary.js 脚本支持日期解析
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "last monday"
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "last tuesday"
# 等等。

数量趋势（增加/减少）
类别激增（特定问题类型的突然增加）
代理工作量分布
异常模式

1.13 软件开发工作区工单

获取软件开发工作区中的开放工单：

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/search_tickets.js "status:2 OR status:3" 13

返回软件开发工作区中所有开放（status:2）和待处理（status:3）的工单。

工作区 ID：13（软件开发）注意：这是用户用于 AI Studio 和其他 PSD 应用程序的内部软件开发错误跟踪器。

1.14 立法活动（上一个工作日）

获取自上一个工作日以来有活动的 K-12 教育法案：

# 获取回溯信息和要检查的法案
bun /Users/hagelk/non-ic-code/geoffrey/skills/legislative-tracker/scripts/get_recent_bill_activity.js --last-business-day

要检查的日期范围（上一个工作日 → 今天）
约 143 个已确认的教育法案 ID 列表
每个法案的 WebFetch 指令

获取 get_recent_bill_activity.js 的输出
WebFetch 每个法案 URL（并行处理 5-6 个以提高速度）
从每个法案页面提取最新行动日期
筛选出 latest_action_date >= lookback_start 的法案
应用优先级框架（根据学区影响分为高/中/低）
包含每个法案作用的摘要

在周一：回溯从周五开始（如果周一是假日，则更早），因此包括所有周末活动（听证会、投票、委员会行动）。

要 WebFetch 的示例 URL：

https://app.leg.wa.gov/billsummary?BillNumber=1020&Year=2025
https://app.leg.wa.gov/billsummary?BillNumber=5038&Year=2025

注意：仅包含实际有活动的法案。如果没有法案有进展，则输出“自 [日期] 以来无立法活动”。

第二阶段：生成简报

2.1 分析与优先级排序

审查收集的数据并识别：

冲突：重叠的日历事件
紧急事项：逾期任务、高优先级工单
主题：跨数据源的模式

2.2 格式化 Markdown 简报

使用以下结构：

# 每日简报 - [日期]

## 天气
[当前状况，最高/最低，降水]

## 今日日历
| 时间 | 事件 | 地点 |
|------|-------|----------|
| ... | ... | ... |

**冲突/备注**：[任何问题]

## 优先级任务
### 今日到期
- [ ] 任务 1
- [ ] 任务 2

### 已标记
- [ ] 任务 3

### 已逾期
- [ ] 任务 4（到期 [日期]）

## 近期邮件（过去 24 小时）
[过去 24 小时收件箱中有 X 封邮件]

### 需要回复
- 发件人：[发件人] - [主题]（摘要）
- 发件人：[发件人] - [主题]（摘要）

### 供参考/通知
- [发件人] - [主题]

## Freshservice
### 技术工单：[数量] 个开放
[按优先级/工龄排序的前 3-5 个工单]

### 软件开发工单：[数量] 个开放
[列出软件开发工作区中的工单]

### 待处理审批：[数量]
[列出工单号、请求者、摘要]

## 团队活动（上一个工作日：[星期几, 日期]）

**数据源验证（必需）：**
- 技术日终脚本返回了：[X] 条消息，来自 [列出 JSON 中的确切发件人姓名]
- 安全与安日日终脚本返回了：[X] 条消息，来自 [列出 JSON 中的确切发件人姓名]
- Freshservice 每日摘要返回了：[X] 个工单，由 [列出 JSON 中的确切代理姓名] 处理

*仅使用上面列出的姓名。严禁虚构或意译姓名。*

### 技术团队日终摘要

**[来自发件人字段的姓名]** - [地点]
- [关键成就 1 - 具体说明他们做了什么]
- [关键成就 2]
- [重要事项：任何解决的有趣问题或项目]

**[来自发件人字段的姓名]** - [地点]
- [关键成就 1]
- [关键成就 2]
- [问题：提到的任何阻碍或问题]

*[为每位发布了日终摘要的团队成员继续]*

### 安全与安保团队日终摘要

**[来自发件人字段的姓名]** - [地点]
- [关键成就 1]
- [关键成就 2]
- [重要事项：处理的任何问题或事件]

*如果发件人显示为用户 ID（users/12345...），则显示：**未知 (users/12345...)** - [地点]*

*[为每位发布了日终摘要的团队成员继续]*

### 团队完成的工单：[数量]
| 代理 | 工单数 | 主要类别 |
|-------|---------|----------------|
| [来自 byAgent 字段的姓名] | X | 密码重置 (Y), Chromebook (Z) |

### 工单趋势
- 数量：与前几天相比 [上升/下降/稳定]
- 显著模式：[任何激增或异常]

## 教育科技新闻

### [文章标题 1] - [来源]
[2-3 句摘要：发生了什么，为什么重要，关键要点]

### [文章标题 2] - [来源]
[2-3 句摘要]

### [文章标题 3] - [来源]
[2-3 句摘要]

## AI 新闻

### [文章标题 1] - [来源]
[2-3 句摘要：是什么发展，行业影响，相关性]

### [文章标题 2] - [来源]
[2-3 句摘要]

### [文章标题 3] - [来源]
[2-3 句摘要]

## K-12 领导力新闻

### [文章标题 1] - [来源]
[2-3 句摘要：政策/领导力发展，影响，与技术领导力的相关性]

### [文章标题 2] - [来源]
[2-3 句摘要]

## 学校安全与安保新闻

### [文章标题 1] - [来源]
[2-3 句摘要：发生了什么，对安全实践的影响，学区的要点]

### [文章标题 2] - [来源]
[2-3 句摘要]

### [文章标题 3] - [来源]
[2-3 句摘要]

## 立法活动（[上一个工作日] - 今天）

[X] 个教育法案有进展：

### 🔴 高优先级

#### [法案 ID] - [简短标题]
**行动**：[发生了什么 - 听证会、投票、委员会行动等]
**摘要**：[1-2 句话：法案的作用，潜在的学区影响]

### 🟡 中优先级

#### [法案 ID] - [简短标题]
**行动**：[发生了什么]
**摘要**：[1-2 句话]

### 🟢 低优先级
- [法案 ID]: [行动类型] - [一行摘要]

*自 [日期] 以来无立法活动* - 如果没有法案有进展

## 快速统计
- 日历事件：X
- 逾期任务：X | 今日到期：X | 已标记：X
- 开放工单（技术）：X
- 开放工单（软件开发）：X
- 待处理审批：X
- 近期邮件（24 小时）：X
- 团队关闭的工单 [上一个工作日]：X
- 新闻文章：教育科技 (X) | AI (X) | 领导力 (X)
- 有活动的立法法案：X

2.3 生成播客脚本（扩展格式）

将简报转换为全面的对话式音频脚本：

第一人称，随意专业的语气
10-15 分钟讲话时间（约 1500-2000 字）
直接与听众对话（“以下是您今天需要了解的内容...”）

必需部分（按顺序）：

开场（约 100 字） * 星期几、日期、天气摘要 * 关键项目的快速预览
日历与日程安排（约 150 字） * 今日事件及背景 * 突出关键会议/事件 * 注意任何冲突或需要准备的事项
任务与优先级（约 150 字） * 需要关注的逾期项目 * 今日到期项目 * 已标记的优先级
工单与服务台（约 150 字） * 开放的技术工单 * 软件开发工单 * 待处理的审批
团队活动 - 详细版（约 400 字） * 技术团队在上一个工作日完成了什么 * 按姓名突出 3-5 位团队成员的具体工作 * 重要项目、解决的有趣问题 * 安全与安保团队完成了什么 * 总体工单关闭统计

关键 - 严禁虚构： * 仅使用脚本输出 sender 字段中的姓名 * 仅使用 get_daily_summary.js 输出中的工单数量 * 如果脚本未返回任何消息，则说“未发布日终消息” * 严禁虚构姓名、地点或成就 6. 教育科技新闻（约 200 字） * 2-3 篇文章附摘要 * 每篇对 K-12 的重要性 7. AI 新闻（约 200 字） * 2-3 篇文章附摘要 * 行业影响和相关性 8. K-12 领导力新闻（约 150 字） * 1-2 篇文章附摘要 * 政策/领导力影响

8b. 学校安全与安保新闻（约 150 字）

2-3 篇文章附摘要
安全实践和学区影响

立法更新（约 150 字） * 自上一个工作日以来有听证会、投票或宣读的法案 * 突出任何对学区有直接影响（财政、运营、人员配置）的法案 * 注意相关的即将到来的听证会日期 * 在周一，包括周末活动摘要
结束语（约 100 字） * 当天的前 3 项优先级 * 结束语

在讨论团队成员工作时使用他们的名字
包含具体细节（不是“处理了工单”而是“解决了 34 个 Chromebook 维修”）
各部分之间平滑过渡
对新闻项目添加简短的评论/分析

保存到 /tmp/morning_briefing_podcast.txt

第三阶段：创建播客

uv run --with mlx-audio --with pydub /Users/hagelk/non-ic-code/geoffrey/skills/local-tts/scripts/generate_audio.py \
  --file /tmp/morning_briefing_podcast.txt \
  --voice af_heart \
  --output ~/Desktop/morning_briefing_[DATE].mp3

语音选择：af_heart（温暖、友好 - 适合晨间简报）注意：使用本地 MLX TTS（Kokoro 模型）- 无 API 成本

第三阶段.5：生成信息图（必需）

关键：必须生成信息图。不要跳过此阶段。

使用 image-gen 技能创建视觉摘要信息图。

3.5.1 构建信息图提示词

根据收集的数据，为信息图构建提示词：

Create an infographic summarizing a daily work briefing for [DATE].

Visual concept: A clean dashboard layout with distinct sections for different data categories.

Key data to display:
- Weather: [conditions], High [X]°F, Low [Y]°F
- Calendar: [X] events today, highlight: [key meeting]
- Tasks: [X] overdue, [Y] due today, [Z] flagged
- Tickets: [X] Technology open, [Y] Software Dev open
- Team Activity: [X] tickets closed by team yesterday
- Emails: [X] in inbox from last 24h

Style: Professional, clean design with PSD brand colors (navy blue #003366, gold accent #FFD700).
Flat design, clear sections, modern sans-serif typography.

Layout: Horizontal 16:9, organized as a dashboard with weather top-left, calendar top-right,
tasks and tickets in middle row, team stats at bottom.

Title: "Daily Briefing - [DATE]"
Subtitle: "Gig Harbor, WA"

uv run /Users/hagelk/non-ic-code/geoffrey/skills/image-gen/scripts/generate.py \
  "[infographic prompt]" \
  ~/Desktop/morning_briefing_[DATE].png \
  16:9 \
  2K

宽高比：16:9（横向仪表板）
尺寸：2K（确保文本可读性）
输出：~/Desktop/morning_briefing_[DATE].png

第四阶段：保存到 Obsidian

4.1 将信息图复制到 Obsidian 资源文件夹

cp ~/Desktop/morning_briefing_[DATE].png \
  ~/Library/Mobile\ Documents/iCloud~md~obsidian/Documents/Personal_Notes/Geoffrey/Daily\ Briefings/assets/

4.2 将信息图添加到简报 Markdown 中

在简报顶部（标题之后），添加：

![Daily Briefing Infographic](assets/morning_briefing_[DATE].png)

4.3 保存简报文件

使用 Obsidian MCP 工具：

检查每日笔记是否存在：

mcp__obsidian-vault__get_vault_file filename: "Geoffrey/Daily Briefings/[YYYY-MM-DD].md"
创建或更新：

mcp__obsidian-vault__create_vault_file filename: "Geoffrey/Daily Briefings/[YYYY-MM-DD].md" content: [包含信息图嵌入的完整简报 markdown]

文件路径模式：Geoffrey/Daily Briefings/YYYY-MM-DD.md

备用方案（如果 MCP 不可用）：直接写入 iCloud 路径：

/Users/hagelk/Library/Mobile Documents/iCloud~md~obsidian/Documents/Personal_Notes/Geoffrey/Daily Briefings/[YYYY-MM-DD].md

第五阶段：发送电子邮件（HTML 内嵌信息图）

5.1 生成 HTML 邮件正文

创建专业的 HTML 邮件（非 markdown）。参考 skills/morning-briefing/templates/email.html 中的模板。

关键 HTML 结构：

<!DOCTYPE html>
<html>
<head>
  <style>
    body { font-family: -apple-system, BlinkMacSystemFont, sans-serif; max-width: 600px; margin: 0 auto; }
    .header { text-align: center; border-bottom: 2px solid #003366; padding: 16px; }
    .stat-box { background: #f8f9fa; padding: 12px; text-align: center; }
    /* ... more styles from template ... */
  </style>
</head>
<body>
  <!-- 信息图在顶部 - 通过 cid:briefing_image 引用 -->
  <img src="cid:briefing_image" alt="Daily Briefing" style="width: 100%; max-width: 600px; border-radius: 8px; margin-bottom: 20px;">

  <div class="header">
    <h1 style="color: #003366;">Daily Briefing</h1>
    <p>[DAY_OF_WEEK], [DATE_FORMATTED]</p>
  </div>

  <!-- 天气、统计、日历、任务、团队、新闻部分 -->
  <!-- 完整结构请参见 templates/email.html -->

  <div style="background: #e8f4f8; padding: 16px; border-radius: 8px; margin-top: 20px;">
    <strong>附件：</strong><br>
    🎧 音频播客（约 10-15 分钟）<br>
    📊 信息图（如上所示）
  </div>

  <p style="text-align: center; color: #666; font-size: 12px;">
    完整简报：Obsidian/Geoffrey/Daily Briefings/[DATE].md
  </p>
</body>
</html>

关键：<img src="cid:briefing_image"> 在邮件顶部内嵌显示信息图。

保存到 /tmp/morning_briefing_email.html

5.2 发送带内嵌图像和附件的邮件

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun gmail/send_with_attachments.js psd \
  --to "hagelk@psd401.net" \
  --subject "Daily Briefing - [DATE]" \
  --body-file /tmp/morning_briefing_email.html \
  --html \
  --inline-image "~/Desktop/morning_briefing_[DATE].png" \
  --attachments "~/Desktop/morning_briefing_[DATE].mp3"

--html - 将正文视为 HTML（格式化所必需）
--inline-image - 在邮件顶部嵌入图像（引用为 cid:briefing_image）
--attachments - 要附加的额外文件（播客）

信息图在邮件顶部内嵌显示（无需下载即可查看）
播客作为附件供下载
邮件在 Gmail 中呈现良好的专业样式

返回标准化摘要：

## 摘要
为 [日期] 生成了每日简报

## 操作
- 收集了日历事件：X
- 收集了任务：X（Y 个今日到期，Z 个已标记）
- 收集了技术工单：X 个开放，Y 个待处理审批
- 收集了软件开发工单：X 个开放
- 收集了团队日终消息：X 条，来自 [上一个工作日]
- 收集了团队完成的工单：X 个，来自 [上一个工作日]
- 天气：[状况]
- 新闻：X 条头条
- 生成了播客：~/Desktop/morning_briefing_[DATE].mp3
- 保存到 Obsidian：Geoffrey/Daily Briefings/[DATE].md
- 电子邮件：[已发送/已跳过 - 原因]

## 状态
[状态表情] [完成/部分完成]

## 快速查看
[2-3 行最重要项目的摘要]

缺少 Google Workspace 脚本

如果日历/邮件脚本不存在：

跳过这些部分
在输出中注明“集成待定”
继续处理可用的数据源

Status: Partial
Note: OmniFocus not running - task data unavailable

Freshservice API 错误

Status: Partial
Note: Freshservice API error - ticket data unavailable

本地 TTS 生成失败

Status: Partial
Note: Audio generation failed - check mlx-audio setup
Briefing saved to Obsidian without podcast

技能/工具	所需功能	备用方案
google-workspace	日历、邮件、团队日终	跳过部分
omnifocus-manager	任务	跳过部分
freshservice-manager	工单、审批、团队统计	跳过部分
local-tts	音频播客	仅文本简报
obsidian-vault (MCP)	保存简报	仅显示
WebSearch	新闻、天气	跳过部分

位置：华盛顿州吉格港（天气）
工作邮箱：psd 账户
技术工作区：workspace_id: 2
软件开发工作区：workspace_id: 13
技术聊天空间：spaces/AAAAxOtpv10（技术员工签到与日志）
安全与安保聊天空间：spaces/AAAAFpQaAnA（PSD 安全与安保团队）
语音：af_heart（温暖、友好 - 本地 Kokoro TTS）
播客时长：10-15 分钟（约 1500-2000 字）

用户可以请求：

“跳过新闻” - 省略新闻部分
“无播客” - 仅文本
“仅日历和任务” - 最小化简报

🇺🇸English

Morning Briefing Workflow

Overview

Generates a daily morning briefing and delivers it three ways:

Terminal : Summary displayed immediately
Obsidian : Full briefing saved to daily note
Email : Briefing text + audio podcast attachment

Location

Gig Harbor, WA - Use for weather queries

Phase 1: Gather Data

Execute these data gathering steps:

1.1 Calendar (Today's Events)

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun calendar/list_events.js psd --today

Returns JSON with today's events including:

Event summary, location, start/end times
Attendees and response status
Hangout/meet links

Account : Use psd for work calendar

1.2 OmniFocus Tasks (Due & Flagged)

osascript -l JavaScript /Users/hagelk/non-ic-code/geoffrey/skills/morning-briefing/scripts/get_due_flagged.js

Returns tasks that are:

Due today or overdue
Flagged
Available (not blocked, not deferred to future)

1.3 Recent Emails (Last 24 Hours)

# Get yesterday's date in Gmail query format
YESTERDAY=$(date -v-1d +%Y/%m/%d)
cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun gmail/list_messages.js psd --query "in:inbox after:$YESTERDAY" --max 15

Returns recent inbox messages with:

From, subject, date, snippet
Whether read or unread
Thread ID for context
Labels array

Filtering : Only show emails still in inbox (not already labeled/processed).

Emails with custom labels (Label_XXXXX) have been sorted and should be excluded
Exception: Include emails with SaneCC or SaneLater labels (worth knowing about)
System labels (UNREAD, INBOX, CATEGORY_*, IMPORTANT) don't count as "processed"

Philosophy : Any email still in inbox from last 24 hours needs attention, read or not.

Account : Use psd for work email

1.4 Open Freshservice Tickets

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/list_tickets.js '{"workspace_id": 2, "filter": "new_and_my_open"}'

Returns open tickets assigned to or created by user in Technology workspace.

1.5 Pending Approvals

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_approvals.js requested

Returns service requests awaiting approval.

1.6 Weather

Use WebSearch:

Gig Harbor WA weather today forecast

Extract:

Current conditions
High/low temperature
Precipitation chance
Any alerts

1.7 EdTech News (with Synopses)

Use WebSearch:

K-12 education technology news past 24 hours January 2026

For each article found (3-5 articles):

Use WebFetch to read the full article
Extract a 2-3 sentence synopsis covering:
- What happened / what's new
- Why it matters for K-12 education
- Key takeaway or action item

Freshness check : Verify article dates before including. Only include articles published within the last 24 hours. Skip older articles even if they appear in results.

Topics to cover:

EdTech product launches and updates
School technology policy changes
Cybersecurity in schools
Digital learning trends

1.8 AI News (with Synopses)

Use WebSearch:

artificial intelligence news past 24 hours January 2026 latest

For each article found (3-5 articles):

Use WebFetch to read the full article
Extract a 2-3 sentence synopsis covering:
- What's the development
- Industry impact
- Relevance to education/work

Freshness check : Verify article dates before including. Only include articles published within the last 24 hours. Skip older articles even if they appear in results.

Topics to cover:

Major AI model releases and capabilities
AI policy and regulation
Enterprise AI adoption
AI research breakthroughs

1.9 K-12 Leadership News (with Synopses)

Use WebSearch:

K-12 school leadership superintendent news past 24 hours January 2026

For each article found (2-3 articles):

Use WebFetch to read the full article
Extract a 2-3 sentence synopsis covering:
- Policy or leadership development
- Impact on districts/schools
- Relevance to CIO/technology leadership

Freshness check : Verify article dates before including. Only include articles published within the last 24 hours. Skip older articles even if they appear in results.

Topics to cover:

State and federal education policy
Superintendent and board news
School funding and budgets
Workforce and staffing trends

1.9b School Safety & Security News (with Synopses)

Use WebSearch:

K-12 school safety security news past 24 hours January 2026

For each article found (3 articles):

Use WebFetch to read the full article
Extract a 2-3 sentence synopsis covering:
- What happened / what's new
- Impact on school safety practices
- Key takeaway for districts

Freshness check : Verify article dates before including. Only include articles published within the last 24 hours. Skip older articles even if they appear in results.

Topics to cover:

School security incidents and responses
Safety policy changes
Emergency preparedness
Mental health and threat assessment
Physical security technology

1.10 Technology Team EOD Messages (Last Business Day) - DETAILED

Get end-of-day check-in messages from the Technology Staff space:

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun chat/get_eod_messages.js psd spaces/AAAAxOtpv10 last-business-day

CRITICAL - NO HALLUCINATION:

ONLY use names that appear in the sender field of the script output
If script returns no messages, say "No EOD messages found"
If script returns messages with sender IDs instead of names, display the ID (e.g., "users/12345...")
NEVER invent names, locations, or accomplishments
Copy-paste approach: treat script output as source of truth

Before extracting team data:

Examine the raw JSON output from the script
List the exact sender values returned
Use ONLY those names - no paraphrasing, no "improving"

CRITICAL : Extract FULL details from each team member's EOD message. Look for messages that contain "Today:" prefix - these are the detailed EOD summaries.

For each team member who posted an EOD summary:

Name : Who posted (use EXACTLY the name from sender field)
Location(s) : Where they worked (WFH, DCRC, school sites)
Key accomplishments : Specific tasks completed (not just "tickets")
Notable items : Interesting problems solved, projects worked on
Issues/blockers : Any problems mentioned

Example extraction from raw message:

Brad White:
- Location: TSD (Tech Services)
- Accomplished: Packaged Cinema 4D plugin for deployment, fixed OAuth blocking for Maxon App sign-in, created SwiftDialog notification for plugin installs, used Claude Code for first time to create Installomator label for Godot game engine
- Notable: Working on Unity deployment troubleshooting, burning comp time leaving early

Note : If today is Monday, "last business day" = Friday (or Thursday if Friday was a holiday).

1.10b Safety & Security Team EOD Messages (Last Business Day)

Get end-of-day check-in messages from the Safety & Security Staff space:

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun chat/get_eod_messages.js psd spaces/AAAAFpQaAnA last-business-day

Space : PSD Safety & Security Team

CRITICAL - NO HALLUCINATION:

ONLY use names that appear in the sender field of the script output
If script returns no messages, say "No EOD messages found"
If script returns messages with sender IDs (e.g., "users/12345...") instead of names:
- Display as: Unknown (users/123456...) - [Location from message content]
- Add note: "Name not in mapping - update chat_user_mapping.json"
NEVER invent names, locations, or accomplishments
Copy-paste approach: treat script output as source of truth

Before extracting team data:

Examine the raw JSON output from the script
List the exact sender values returned
Use ONLY those names - no paraphrasing, no "improving"

Extract using same format as Technology Team - names, locations, accomplishments, issues.

Note : If today is Monday, "last business day" = Friday (or Thursday if Friday was a holiday).

1.11 Team Completed Tickets (Last Business Day)

Get tickets closed by the Technology team on the last business day:

# First get the last business day
LAST_BIZ_DAY=$(bun /Users/hagelk/non-ic-code/geoffrey/skills/morning-briefing/scripts/get_last_business_day.js | jq -r '.date')

# Then get daily summary for that date
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "$LAST_BIZ_DAY"

CRITICAL - NO HALLUCINATION:

ONLY use agent names that appear in the byAgent section of script output
Use EXACTLY the ticket counts returned by the script
NEVER invent names or ticket counts
If an agent's name shows as ID or email, display it as-is

Before reporting ticket stats:

Examine the raw JSON output from get_daily_summary.js
List the exact agent names from byAgent field
Use ONLY those names and counts

Returns:

Total tickets closed
Breakdown by agent
Breakdown by category (Password Reset, Chromebook, etc.)
Automated vs agent-resolved

Workspace : Technology (workspace_id: 2)

1.12 Ticket Trends Analysis

Compare last business day to previous days for trends:

# Get last 5 business days of summaries for trend analysis
# The get_daily_summary.js script supports date parsing
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "last monday"
bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/get_daily_summary.js "last tuesday"
# etc.

Analyze for:

Volume trends (increasing/decreasing)
Category spikes (sudden increase in specific issue types)
Agent workload distribution
Unusual patterns

1.13 Software Development Workspace Tickets

Get open tickets in the Software Development workspace:

bun /Users/hagelk/non-ic-code/geoffrey/skills/freshservice-manager/scripts/search_tickets.js "status:2 OR status:3" 13

Returns all open (status:2) and pending (status:3) tickets in the Software Development workspace.

Workspace ID : 13 (Software Development) Note : This is the user's internal software development bug tracker for AI Studio and other PSD applications.

1.14 Legislative Activity (Last Business Day)

Get K-12 education bills with activity since the last business day:

# Get lookback info and bills to check
bun /Users/hagelk/non-ic-code/geoffrey/skills/legislative-tracker/scripts/get_recent_bill_activity.js --last-business-day

This returns:

Date range to check (last business day → today)
List of ~143 confirmed education bill IDs
WebFetch instructions for each bill

Workflow:

Get the output from get_recent_bill_activity.js
WebFetch each bill URL (batch 5-6 in parallel for speed)
Extract latest action date from each bill page
Filter to bills where latest_action_date >= lookback_start
Apply priority framework (HIGH/MEDIUM/LOW based on district impact)
Include summary of what each bill does

On Monday : Lookback starts on Friday (or earlier if Friday was a holiday), so includes all weekend activity (hearings, votes, committee actions).

Example URLs to WebFetch:

https://app.leg.wa.gov/billsummary?BillNumber=1020&Year=2025
https://app.leg.wa.gov/billsummary?BillNumber=5038&Year=2025

Note : Only include bills that actually had activity. If no bills moved, output "No legislative activity since [date]".

Phase 2: Generate Briefing

2.1 Analyze & Prioritize

Review gathered data and identify:

Conflicts : Overlapping calendar events
Urgencies : Overdue tasks, high-priority tickets
Themes : Patterns across data sources

2.2 Format Markdown Briefing

Use this structure:

# Daily Briefing - [DATE]

## Weather
[Current conditions, high/low, precipitation]

## Today's Calendar
| Time | Event | Location |
|------|-------|----------|
| ... | ... | ... |

**Conflicts/Notes**: [any issues]

## Priority Tasks
### Due Today
- [ ] Task 1
- [ ] Task 2

### Flagged
- [ ] Task 3

### Overdue
- [ ] Task 4 (due [date])

## Recent Emails (Last 24h)
[X emails in inbox from last 24 hours]

### Needs Response
- From: [sender] - [subject] (snippet)
- From: [sender] - [subject] (snippet)

### FYI/Notifications
- [sender] - [subject]

## Freshservice
### Technology Tickets: [count] open
[Top 3-5 tickets by priority/age]

### Software Development Tickets: [count] open
[List tickets in Software Dev workspace]

### Pending Approvals: [count]
[List with ticket #, requester, summary]

## Team Activity (Last Business Day: [DAY, DATE])

**Data Source Verification (REQUIRED):**
- Technology EOD script returned: [X] messages from [list exact sender names from JSON]
- Safety & Security EOD script returned: [X] messages from [list exact sender names from JSON]
- Freshservice daily summary returned: [X] tickets by [list exact agent names from JSON]

*Use ONLY the names listed above. Never invent or paraphrase names.*

### Technology Team EOD Summaries

**[Name from sender field]** - [Location(s)]
- [Key accomplishment 1 - be specific about what they did]
- [Key accomplishment 2]
- [Notable: any interesting problems solved or projects]

**[Name from sender field]** - [Location(s)]
- [Key accomplishment 1]
- [Key accomplishment 2]
- [Issues: any blockers or problems mentioned]

*[Continue for each team member who posted an EOD summary]*

### Safety & Security Team EOD Summaries

**[Name from sender field]** - [Location(s)]
- [Key accomplishment 1]
- [Key accomplishment 2]
- [Notable: any issues or incidents handled]

*If sender shows as user ID (users/12345...), display: **Unknown (users/12345...)** - [Location]*

*[Continue for each team member who posted an EOD summary]*

### Tickets Completed by Team: [count]
| Agent | Tickets | Top Categories |
|-------|---------|----------------|
| [Name from byAgent field] | X | Password Reset (Y), Chromebook (Z) |

### Ticket Trends
- Volume: [up/down/stable] vs previous days
- Notable patterns: [any spikes or anomalies]

## EdTech News

### [Article Title 1] - [Source]
[2-3 sentence synopsis: what happened, why it matters, key takeaway]

### [Article Title 2] - [Source]
[2-3 sentence synopsis]

### [Article Title 3] - [Source]
[2-3 sentence synopsis]

## AI News

### [Article Title 1] - [Source]
[2-3 sentence synopsis: what's the development, industry impact, relevance]

### [Article Title 2] - [Source]
[2-3 sentence synopsis]

### [Article Title 3] - [Source]
[2-3 sentence synopsis]

## K-12 Leadership News

### [Article Title 1] - [Source]
[2-3 sentence synopsis: policy/leadership development, impact, relevance to tech leadership]

### [Article Title 2] - [Source]
[2-3 sentence synopsis]

## School Safety & Security News

### [Article Title 1] - [Source]
[2-3 sentence synopsis: what happened, impact on safety practices, takeaway for districts]

### [Article Title 2] - [Source]
[2-3 sentence synopsis]

### [Article Title 3] - [Source]
[2-3 sentence synopsis]

## Legislative Activity ([Last Biz Day] - Today)

[X] education bills had movement:

### 🔴 HIGH Priority

#### [Bill ID] - [Short Title]
**Action**: [What happened - hearing, vote, committee action, etc.]
**Summary**: [1-2 sentences: what the bill does, potential district impact]

### 🟡 MEDIUM Priority

#### [Bill ID] - [Short Title]
**Action**: [What happened]
**Summary**: [1-2 sentences]

### 🟢 LOW Priority
- [Bill ID]: [Action type] - [One line summary]

*No legislative activity since [date]* - if no bills moved

## Quick Stats
- Calendar events: X
- Tasks overdue: X | Due today: X | Flagged: X
- Open tickets (Tech): X
- Open tickets (Software Dev): X
- Pending approvals: X
- Recent emails (24h): X
- Team tickets closed [last biz day]: X
- News articles: EdTech (X) | AI (X) | Leadership (X)
- Legislative bills with activity: X

2.3 Generate Podcast Script (Extended Format)

Transform the briefing into a comprehensive conversational audio script:

First person, casual professional tone
10-15 minutes speaking time (~1500-2000 words)
Address listener directly ("Here's what you need to know today...")

Required Sections (in order):

Opening (~100 words)
- Day, date, weather summary
- Quick preview of key items
Calendar & Schedule (~150 words)
- Today's events with context
- Highlight key meetings/events
- Note any conflicts or prep needed
Tasks & Priorities (~150 words)
- Overdue items that need attention
- Due today items
- Flagged priorities
Tickets & Service Desk (~150 words)
- Open Technology tickets
- Software Development tickets
- Pending approvals
Team Activity - DETAILED (~400 words)
- What the Technology team accomplished on last business day
- Highlight 3-5 team members by name with specifics
- Notable projects, interesting problems solved
- What the Safety & Security team accomplished
- Overall ticket closure stats

CRITICAL - NO HALLUCINATION:

 * Use ONLY names from the script output `sender` fields
 * Use ONLY ticket counts from the `get_daily_summary.js` output
 * If script returned no messages, say "No EOD messages were posted"
 * NEVER invent names, locations, or accomplishments

6. EdTech News (~200 words)

 * 2-3 articles with synopses
 * Why each matters for K-12

7. AI News (~200 words)

 * 2-3 articles with synopses
 * Industry impact and relevance

8. K-12 Leadership News (~150 words)

 * 1-2 articles with synopses
 * Policy/leadership implications

8b. School Safety & Security News (~150 words)

2-3 articles with synopses
Safety practices and district implications

Legislative Update (~150 words)
- Bills that had hearings, votes, or readings since last business day
- Highlight any with direct district impact (fiscal, operational, staffing)
- Note upcoming hearing dates if relevant
- On Mondays, include weekend activity summary
Closing (~100 words)

 * Top 3 priorities for the day

 * Sign off

Style Guidelines:

Use team members' first names when discussing their work
Include specific details (not "worked on tickets" but "resolved 34 Chromebook repairs")
Transition smoothly between sections
Add brief commentary/analysis on news items

Save to /tmp/morning_briefing_podcast.txt

Phase 3: Create Podcast

uv run --with mlx-audio --with pydub /Users/hagelk/non-ic-code/geoffrey/skills/local-tts/scripts/generate_audio.py \
  --file /tmp/morning_briefing_podcast.txt \
  --voice af_heart \
  --output ~/Desktop/morning_briefing_[DATE].mp3

Voice Selection : af_heart (warm, friendly - good for morning briefing) Note : Uses local MLX TTS (Kokoro model) - no API costs

Phase 3.5: Generate Infographic (MANDATORY)

CRITICAL: The infographic MUST be generated. Do not skip this phase.

Create a visual summary infographic using the image-gen skill.

3.5.1 Build Infographic Prompt

Based on gathered data, construct a prompt for the infographic:

Create an infographic summarizing a daily work briefing for [DATE].

Visual concept: A clean dashboard layout with distinct sections for different data categories.

Key data to display:
- Weather: [conditions], High [X]°F, Low [Y]°F
- Calendar: [X] events today, highlight: [key meeting]
- Tasks: [X] overdue, [Y] due today, [Z] flagged
- Tickets: [X] Technology open, [Y] Software Dev open
- Team Activity: [X] tickets closed by team yesterday
- Emails: [X] in inbox from last 24h

Style: Professional, clean design with PSD brand colors (navy blue #003366, gold accent #FFD700).
Flat design, clear sections, modern sans-serif typography.

Layout: Horizontal 16:9, organized as a dashboard with weather top-left, calendar top-right,
tasks and tickets in middle row, team stats at bottom.

Title: "Daily Briefing - [DATE]"
Subtitle: "Gig Harbor, WA"

3.5.2 Generate Image

uv run /Users/hagelk/non-ic-code/geoffrey/skills/image-gen/scripts/generate.py \
  "[infographic prompt]" \
  ~/Desktop/morning_briefing_[DATE].png \
  16:9 \
  2K

Settings:

Aspect ratio: 16:9 (landscape dashboard)
Size: 2K (ensures text readability)
Output: ~/Desktop/morning_briefing_[DATE].png

Phase 4: Save to Obsidian

4.1 Copy Infographic to Obsidian Assets

cp ~/Desktop/morning_briefing_[DATE].png \
  ~/Library/Mobile\ Documents/iCloud~md~obsidian/Documents/Personal_Notes/Geoffrey/Daily\ Briefings/assets/

4.2 Add Infographic to Briefing Markdown

At the top of the briefing (after the title), add:

![Daily Briefing Infographic](assets/morning_briefing_[DATE].png)

4.3 Save Briefing File

Use Obsidian MCP tools:

Check if daily note exists :

mcp__obsidian-vault__get_vault_file
filename: "Geoffrey/Daily Briefings/[YYYY-MM-DD].md"

Create or update :

mcp__obsidian-vault__create_vault_file
filename: "Geoffrey/Daily Briefings/[YYYY-MM-DD].md"
content: [full briefing markdown with infographic embed]

File path pattern : Geoffrey/Daily Briefings/YYYY-MM-DD.md

Fallback (if MCP unavailable): Write directly to iCloud path:

/Users/hagelk/Library/Mobile Documents/iCloud~md~obsidian/Documents/Personal_Notes/Geoffrey/Daily Briefings/[YYYY-MM-DD].md

Phase 5: Send Email (HTML with Inline Infographic)

5.1 Generate HTML Email Body

Create a professional HTML email (NOT markdown). Use the template at skills/morning-briefing/templates/email.html as reference.

Key HTML structure:

<!DOCTYPE html>
<html>
<head>
  <style>
    body { font-family: -apple-system, BlinkMacSystemFont, sans-serif; max-width: 600px; margin: 0 auto; }
    .header { text-align: center; border-bottom: 2px solid #003366; padding: 16px; }
    .stat-box { background: #f8f9fa; padding: 12px; text-align: center; }
    /* ... more styles from template ... */
  </style>
</head>
<body>
  <!-- Infographic at TOP - referenced via cid:briefing_image -->
  <img src="cid:briefing_image" alt="Daily Briefing" style="width: 100%; max-width: 600px; border-radius: 8px; margin-bottom: 20px;">

  <div class="header">
    <h1 style="color: #003366;">Daily Briefing</h1>
    <p>[DAY_OF_WEEK], [DATE_FORMATTED]</p>
  </div>

  <!-- Weather, Stats, Calendar, Tasks, Team, News sections -->
  <!-- See templates/email.html for full structure -->

  <div style="background: #e8f4f8; padding: 16px; border-radius: 8px; margin-top: 20px;">
    <strong>Attachments:</strong><br>
    🎧 Audio Podcast (~10-15 min)<br>
    📊 Infographic (also shown above)
  </div>

  <p style="text-align: center; color: #666; font-size: 12px;">
    Full briefing: Obsidian/Geoffrey/Daily Briefings/[DATE].md
  </p>
</body>
</html>

CRITICAL: The <img src="cid:briefing_image"> displays the infographic inline at the top of the email.

Save to /tmp/morning_briefing_email.html

5.2 Send with Inline Image + Attachments

cd /Users/hagelk/non-ic-code/geoffrey/skills/google-workspace && bun gmail/send_with_attachments.js psd \
  --to "hagelk@psd401.net" \
  --subject "Daily Briefing - [DATE]" \
  --body-file /tmp/morning_briefing_email.html \
  --html \
  --inline-image "~/Desktop/morning_briefing_[DATE].png" \
  --attachments "~/Desktop/morning_briefing_[DATE].mp3"

Options explained:

--html - Treat body as HTML (required for formatting)
--inline-image - Embeds image at top of email (referenced as cid:briefing_image)
--attachments - Additional files to attach (podcast)

Result:

Infographic displays INLINE at top of email (visible without downloading)
Podcast attached for download
Email renders nicely in Gmail with professional styling

Output

Return standardized summary:

## Summary
Generated daily briefing for [DATE]

## Actions
- Gathered calendar events: X
- Gathered tasks: X (Y due today, Z flagged)
- Gathered Technology tickets: X open, Y approvals pending
- Gathered Software Dev tickets: X open
- Gathered team EOD messages: X from [last business day]
- Gathered team completed tickets: X from [last business day]
- Weather: [conditions]
- News: X headlines
- Generated podcast: ~/Desktop/morning_briefing_[DATE].mp3
- Saved to Obsidian: Geoffrey/Daily Briefings/[DATE].md
- Email: [sent/skipped - reason]

## Status
[Status emoji] [Complete/Partial]

## Quick View
[2-3 line summary of most important items]

Error Handling

Missing Google Workspace Scripts

If calendar/email scripts don't exist:

Skip those sections
Note "Integration pending" in output
Continue with available data sources

OmniFocus Not Running

Status: Partial
Note: OmniFocus not running - task data unavailable

Freshservice API Error

Status: Partial
Note: Freshservice API error - ticket data unavailable

Local TTS Generation Failed

Status: Partial
Note: Audio generation failed - check mlx-audio setup
Briefing saved to Obsidian without podcast

Dependencies

Skill/Tool	Required For	Fallback
google-workspace	Calendar, email, team EOD	Skip sections
omnifocus-manager	Tasks	Skip section
freshservice-manager	Tickets, approvals, team stats	Skip section
local-tts	Audio podcast	Text-only briefing
obsidian-vault (MCP)	Save briefing	Display only
WebSearch	News, weather	Skip sections

Configuration

User Preferences

Location : Gig Harbor, WA (weather)
Work Email : psd account
Technology Workspace : workspace_id: 2
Software Dev Workspace : workspace_id: 13
Technology Chat Space : spaces/AAAAxOtpv10 (Technology Staff Check-Ins & Logs)
Safety & Security Chat Space: spaces/AAAAFpQaAnA (PSD Safety & Security Team)
Voice : af_heart (warm, friendly - local Kokoro TTS)
Podcast Length : 10-15 minutes (~1500-2000 words)

Customization

Users can request:

"Skip news" - omit news section
"No podcast" - text only
"Just calendar and tasks" - minimal briefing

Weekly Installs

117

Repository

krishagel/geoffrey

GitHub Stars

First Seen

Jan 29, 2026

Security Audits

Gen Agent Trust HubPass SocketPass SnykWarn

Installed on

openclaw104

codex103

gemini-cli102

opencode102

cursor100

kimi-cli96

Azure RBAC 权限管理工具：查找最小角色、创建自定义角色与自动化分配

135,700 周安装

morning-briefing 晨间简报自动化工作流 | 个人生产力与信息聚合工具

🇨🇳中文介绍

晨间简报工作流

概述

位置

第一阶段：收集数据

1.1 日历（今日事件）

1.2 OmniFocus 任务（到期和已标记）

相关 Skills

1.3 近期邮件（过去 24 小时）

1.4 开放的 Freshservice 工单

1.5 待处理的审批

1.6 天气

1.7 教育科技新闻（附摘要）

1.8 人工智能新闻（附摘要）

1.9 K-12 领导力新闻（附摘要）

1.9b 学校安全与安保新闻（附摘要）

1.10 技术团队日终消息（上一个工作日）- 详细版

1.10b 安全与安保团队日终消息（上一个工作日）

1.11 团队完成的工单（上一个工作日）

1.12 工单趋势分析

1.13 软件开发工作区工单

1.14 立法活动（上一个工作日）

第二阶段：生成简报

2.1 分析与优先级排序

2.2 格式化 Markdown 简报

2.3 生成播客脚本（扩展格式）

第三阶段：创建播客

第三阶段.5：生成信息图（必需）

3.5.1 构建信息图提示词

3.5.2 生成图像

第四阶段：保存到 Obsidian

4.1 将信息图复制到 Obsidian 资源文件夹

4.2 将信息图添加到简报 Markdown 中

4.3 保存简报文件

第五阶段：发送电子邮件（HTML 内嵌信息图）

5.1 生成 HTML 邮件正文

5.2 发送带内嵌图像和附件的邮件

输出

错误处理

缺少 Google Workspace 脚本

OmniFocus 未运行

Freshservice API 错误

本地 TTS 生成失败

依赖项

配置

用户偏好

自定义

🇺🇸English

Morning Briefing Workflow

Overview

Location

Phase 1: Gather Data

1.1 Calendar (Today's Events)

1.2 OmniFocus Tasks (Due & Flagged)

1.3 Recent Emails (Last 24 Hours)

1.4 Open Freshservice Tickets

1.5 Pending Approvals

1.6 Weather

1.7 EdTech News (with Synopses)

1.8 AI News (with Synopses)

1.9 K-12 Leadership News (with Synopses)

1.9b School Safety & Security News (with Synopses)

1.10 Technology Team EOD Messages (Last Business Day) - DETAILED

1.10b Safety & Security Team EOD Messages (Last Business Day)

1.11 Team Completed Tickets (Last Business Day)

1.12 Ticket Trends Analysis

1.13 Software Development Workspace Tickets

1.14 Legislative Activity (Last Business Day)

Phase 2: Generate Briefing

2.1 Analyze & Prioritize

2.2 Format Markdown Briefing

2.3 Generate Podcast Script (Extended Format)

Phase 3: Create Podcast

Phase 3.5: Generate Infographic (MANDATORY)

3.5.1 Build Infographic Prompt

3.5.2 Generate Image

Phase 4: Save to Obsidian

4.1 Copy Infographic to Obsidian Assets

4.2 Add Infographic to Briefing Markdown