Back to Details

openclaw-executive-assistant-workshop

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

OpenClaw Executive Assistant Workshop

OpenClaw AI执行助理工作坊

Skill by ara.so — Hermes Skills collection.
This workshop teaches you to build local-first AI executive assistant workflows using OpenClaw. All operations use local files, markdown outputs, and copy/paste prompts—no live integrations required.
由ara.so提供的技能——Hermes技能合集。
本工作坊将教你使用OpenClaw构建本地优先的AI执行助理工作流。所有操作均使用本地文件、Markdown输出和复制粘贴提示词——无需实时集成。

What This Project Does

项目功能

The OpenClaw Executive Assistant Workshop provides three core automation patterns:
  1. Data Intake Review — Turn unknown files into trustworthy intake reports
  2. Operational Memory — Turn work residue into daily logs and weekly summaries
  3. Offline Communications Triage — Turn exported emails into action lists
All workflows produce reviewable markdown artifacts that stay in local folders.
OpenClaw AI执行助理工作坊提供三种核心自动化模式：
  1. 数据接收审核 — 将未知文件转换为可信的接收报告
  2. 操作记忆 — 将工作遗留内容转换为每日日志和每周总结
  3. 离线通讯分类 — 将导出的邮件转换为行动清单
所有工作流生成的可审核Markdown产物均保存在本地文件夹中。

Repository Structure

仓库结构

.
├── webinar-runbook.html          # Main walkthrough guide
└── code-along/
    ├── INDEX.md                   # Overview
    ├── 01-data-intake-review/     # Exercise 1
    ├── 02-operational-memory/     # Exercise 2
    ├── 03-offline-communications-triage/  # Exercise 3
    └── mission-control/           # Optional dashboard
.
├── webinar-runbook.html          # 主演练指南
└── code-along/
    ├── INDEX.md                   # 概述
    ├── 01-data-intake-review/     # 练习1
    ├── 02-operational-memory/     # 练习2
    ├── 03-offline-communications-triage/  # 练习3
    └── mission-control/           # 可选仪表盘

Installation

安装步骤

bash
git clone https://github.com/dandenney/webinars-build-your-own-executive-assistant-with-openclaw.git
cd webinars-build-your-own-executive-assistant-with-openclaw
Open 
webinar-runbook.html
in your browser for the full walkthrough.
bash
git clone https://github.com/dandenney/webinars-build-your-own-executive-assistant-with-openclaw.git
cd webinars-build-your-own-executive-assistant-with-openclaw
在浏览器中打开
webinar-runbook.html
查看完整演练流程。

Exercise 1: Data Intake Review

练习1：数据接收审核

Goal: Turn unknown files into a trustworthy intake report.
目标： 将未知文件转换为可信的接收报告。

Folder Structure

文件夹结构

code-along/01-data-intake-review/
├── incoming/               # Files to inspect
├── prompts/
│   └── intake-review.md   # Prompt template
├── outputs/
│   └── intake-review.md   # Generated report
└── expected/
    └── report-outline.md  # Expected format
code-along/01-data-intake-review/
├── incoming/               # 待检查文件
├── prompts/
│   └── intake-review.md   # 提示词模板
├── outputs/
│   └── intake-review.md   # 生成的报告
└── expected/
    └── report-outline.md  # 预期格式

Workflow

工作流程

  1. Place unknown files in 
    incoming/
  2. Open 
    prompts/intake-review.md
    and copy the prompt
  3. Provide the prompt to your AI assistant along with the files from 
    incoming/
  4. Save the generated report to 
    outputs/intake-review.md
  1. 将未知文件放入
    incoming/
    文件夹
  2. 打开
    prompts/intake-review.md
    并复制提示词
  3. 将提示词与
    incoming/
    中的文件一起提供给你的AI助理
  4. 将生成的报告保存到
    outputs/intake-review.md

Example Prompt Pattern

示例提示词模板

markdown
undefined
markdown
undefined

Data Intake Review Prompt

数据接收审核提示词

Review all files in the incoming folder and generate a structured report:
审核incoming文件夹中的所有文件并生成结构化报告：

Files Reviewed

已审核文件

  • List each file with type, size, and brief description
  • 列出每个文件的类型、大小和简要描述

Content Summary

内容摘要

  • Summarize key information found
  • 总结发现的关键信息

Recommended Actions

建议行动

  • Suggest next steps for each file
  • 为每个文件建议后续步骤

Risk Assessment

风险评估

  • Flag any suspicious or concerning content
Output format: Markdown
undefined
  • 标记任何可疑或需要关注的内容
输出格式：Markdown
undefined

Expected Output Structure

预期输出结构

markdown
undefined
markdown
undefined

Intake Review Report

接收审核报告

Generated: YYYY-MM-DD
生成时间：YYYY-MM-DD

Files Reviewed

已审核文件

  • document.pdf
    (245 KB) - Project proposal
  • data.csv
    (12 KB) - Contact list
  • notes.txt
    (3 KB) - Meeting notes
  • document.pdf
    (245 KB) - 项目提案
  • data.csv
    (12 KB) - 联系人列表
  • notes.txt
    (3 KB) - 会议记录

Content Summary

内容摘要

[Structured summary of findings]
[结构化的发现总结]

Recommended Actions

建议行动

  1. document.pdf: Review and file in projects folder
  2. data.csv: Import to CRM system
  3. notes.txt: Add to daily log
  1. document.pdf：审核后存入项目文件夹
  2. data.csv：导入CRM系统
  3. notes.txt：添加到每日日志

Risk Assessment

风险评估

No security concerns identified.
undefined
未发现安全问题。
undefined

Exercise 2: Operational Memory

练习2：操作记忆

Goal: Turn work residue into daily logs and weekly summaries.
目标： 将工作遗留内容转换为每日日志和每周总结。

Folder Structure

文件夹结构

code-along/02-operational-memory/
├── inbox/                 # Notes and work residue
├── prompts/
│   ├── daily-log.md      # Daily prompt
│   └── weekly-hype.md    # Weekly prompt
├── outputs/
│   ├── daily-log.md
│   └── weekly-hype.md
└── schedule/
    ├── cron-examples.md
    └── heartbeat-note.md
code-along/02-operational-memory/
├── inbox/                 # 笔记和工作遗留内容
├── prompts/
│   ├── daily-log.md      # 每日提示词
│   └── weekly-hype.md    # 每周提示词
├── outputs/
│   ├── daily-log.md
│   └── weekly-hype.md
└── schedule/
    ├── cron-examples.md
    └── heartbeat-note.md

Daily Log Workflow

每日日志工作流程

markdown
undefined
markdown
undefined

Daily Log Prompt

每日日志提示词

Review all notes in the inbox folder from today and generate:
审核今日inbox文件夹中的所有笔记并生成：

What Got Done

已完成事项

  • Completed tasks and accomplishments
  • 已完成的任务和成果

What's In Progress

进行中事项

  • Ongoing work with current status
  • 正在进行的工作及当前状态

Blockers & Questions

障碍与问题

  • Issues requiring attention
  • 需要关注的问题

Tomorrow's Priority

明日优先级

  • Top 3 items to focus on
Use bullet points. Keep it concise.
undefined
  • 需重点关注的前3项任务
使用项目符号列表，保持简洁。
undefined

Weekly Summary Workflow

每周总结工作流程

markdown
undefined
markdown
undefined

Weekly Hype Prompt

每周总结提示词

Review daily logs from the past week and generate:
审核过去一周的每日日志并生成：

Week at a Glance

本周概览

  • High-level summary of the week
  • 本周的高层总结

Key Wins

关键成果

  • Major accomplishments
  • 主要成就

Progress Metrics

进度指标

  • Quantifiable progress indicators
  • 可量化的进度指标

Next Week's Focus

下周重点

  • Top priorities for the coming week
  • 下周的首要任务

Learnings & Insights

经验与见解

  • Key takeaways
Tone: Positive and momentum-building.
undefined
  • 关键收获
语气：积极且富有动力。
undefined

Automation Example (Cron)

自动化示例（Cron）

bash
undefined
bash
undefined

Daily log generation at 5 PM

每天下午5点生成每日日志

0 17 * * * cd ~/openclaw-assistant/code-along/02-operational-memory && ./generate-daily-log.sh
0 17 * * * cd ~/openclaw-assistant/code-along/02-operational-memory && ./generate-daily-log.sh

Weekly summary on Friday at 4 PM

每周五下午4点生成每周总结

0 16 * * 5 cd ~/openclaw-assistant/code-along/02-operational-memory && ./generate-weekly-summary.sh
undefined
0 16 * * 5 cd ~/openclaw-assistant/code-along/02-operational-memory && ./generate-weekly-summary.sh
undefined

Example Daily Log Output

每日日志输出示例

markdown
undefined
markdown
undefined

Daily Log - 2026-05-15

每日日志 - 2026-05-15

What Got Done

已完成事项

  • Completed client proposal review
  • Fixed bug in authentication module
  • Responded to 12 priority emails
  • 完成客户提案审核
  • 修复认证模块中的漏洞
  • 回复12封优先级邮件

What's In Progress

进行中事项

  • Dashboard redesign (60% complete)
  • API documentation update
  • Budget planning for Q3
  • 仪表盘重新设计（完成60%）
  • API文档更新
  • Q3预算规划

Blockers & Questions

障碍与问题

  • Need design approval for homepage
  • Waiting on legal review of contract
  • 需获得首页设计批准
  • 等待合同的法律审核

Tomorrow's Priority

明日优先级

  1. Finish dashboard mockups
  2. Deploy authentication fix
  3. Schedule Q3 planning meeting
undefined
  1. 完成仪表盘原型
  2. 部署认证修复方案
  3. 安排Q3规划会议
undefined

Exercise 3: Offline Communications Triage

练习3：离线通讯分类

Goal: Turn exported emails into an action list.
目标： 将导出的邮件转换为行动清单。

Folder Structure

文件夹结构

code-along/03-offline-communications-triage/
├── eml/                   # Exported email files
├── prompts/
│   └── email-triage.md   # Triage prompt
├── outputs/
│   └── email-triage.md   # Generated report
└── expected/
    └── report-outline.md
code-along/03-offline-communications-triage/
├── eml/                   # 导出的邮件文件
├── prompts/
│   └── email-triage.md   # 分类提示词
├── outputs/
│   └── email-triage.md   # 生成的报告
└── expected/
    └── report-outline.md

Email Triage Workflow

邮件分类工作流程

  1. Export emails as 
    .eml
    files to the 
    eml/
    folder
  2. Use the triage prompt to analyze all emails
  3. Generate an action-oriented report
  1. 将邮件导出为
    .eml
    文件并存入
    eml/
    文件夹
  2. 使用分类提示词分析所有邮件
  3. 生成面向行动的报告

Example Triage Prompt

分类提示词示例

markdown
undefined
markdown
undefined

Email Triage Prompt

邮件分类提示词

Analyze all emails in the eml folder and generate:
分析eml文件夹中的所有邮件并生成：

Urgent - Respond Today

紧急 - 今日回复

  • Emails requiring immediate response
  • Include sender, subject, required action
  • 需要立即回复的邮件
  • 包含发件人、主题和所需行动

Important - This Week

重要 - 本周处理

  • Important emails with flexible deadlines
  • Include context and suggested response approach
  • 重要但截止日期灵活的邮件
  • 包含背景信息和建议回复方式

FYI - Read & Archive

仅供参考 - 阅读后归档

  • Informational emails requiring no action
  • 无需行动的信息类邮件

Delegate

委派

  • Emails that should be handled by others
  • 应由他人处理的邮件

Low Priority

低优先级

  • Can be addressed later or ignored
For each email, provide:
  • Sender
  • Subject
  • 1-sentence summary
  • Recommended action
undefined
  • 可稍后处理或忽略的邮件
针对每封邮件，提供：
  • 发件人
  • 主题
  • 1句话摘要
  • 建议行动
undefined

Example Triage Output

分类输出示例

markdown
undefined
markdown
undefined

Email Triage Report - 2026-05-15

邮件分类报告 - 2026-05-15

Urgent - Respond Today

紧急 - 今日回复

1. Client Deliverable Review

1. 客户交付物审核

  • From: sarah@client.com
  • Summary: Requesting feedback on Q2 deliverables
  • Action: Review attached document and send approval by EOD
  • 发件人： sarah@client.com
  • 摘要： 请求对Q2交付物提供反馈
  • 行动： 审核附件文档并在今日下班前发送批准意见

2. Team Meeting Conflict

2. 团队会议冲突

  • From: john@company.com
  • Summary: Schedule conflict for Thursday's planning session
  • Action: Propose alternate time slots
  • 发件人： john@company.com
  • 摘要： 周四规划会议时间冲突
  • 行动： 提议替代时间段

Important - This Week

重要 - 本周处理

3. Budget Approval Request

3. 预算批准请求

  • From: finance@company.com
  • Summary: Q3 budget needs sign-off
  • Action: Review spreadsheet and approve by Friday
  • 发件人： finance@company.com
  • 摘要： Q3预算需要签字确认
  • 行动： 审核电子表格并在周五前批准

FYI - Read & Archive

仅供参考 - 阅读后归档

4. Industry Newsletter

4. 行业通讯

  • From: newsletter@industry.org
  • Summary: Weekly roundup of industry news
  • Action: Skim and archive
  • 发件人： newsletter@industry.org
  • 摘要： 行业新闻每周汇总
  • 行动： 浏览后归档

Delegate

委派

5. Technical Support Request

5. 技术支持请求

  • From: support@client.com
  • Summary: API integration question
  • Action: Forward to development team lead
undefined
  • 发件人： support@client.com
  • 摘要： API集成问题咨询
  • 行动： 转发给开发团队负责人
undefined

Configuration Patterns

配置模式

Environment Setup

环境设置

Create a 
.env
file for any API configurations (if extending beyond local-only):
bash
undefined
创建
.env
文件用于存储API配置（如果扩展到本地以外的功能）：
bash
undefined

.env

.env

OPENCLAW_WORKSPACE=~/openclaw-assistant/code-along OUTPUT_FORMAT=markdown TIMEZONE=America/New_York
undefined
OPENCLAW_WORKSPACE=~/openclaw-assistant/code-along OUTPUT_FORMAT=markdown TIMEZONE=America/New_York
undefined

Prompt Template Variables

提示词模板变量

Structure prompts with consistent sections:
markdown
undefined
使用一致的结构组织提示词：
markdown
undefined

[Prompt Name]

[提示词名称]

Input: [Description of input files/data] Output: [Description of expected output] Tone: [Formal/Casual/Technical]
输入： [输入文件/数据描述] 输出： [预期输出描述] 语气： [正式/随意/技术]

Instructions

说明

[Detailed steps]
[详细步骤]

Output Format

输出格式

[Markdown structure template]
undefined
[Markdown结构模板]
undefined

Common Patterns

通用模式

1. Batch File Processing

1. 批量文件处理

bash
#!/bin/bash
bash
#!/bin/bash

Process all files in a directory

处理目录中的所有文件

INPUT_DIR="./incoming" OUTPUT_FILE="./outputs/report.md"
echo "# Batch Processing Report" > "$OUTPUT_FILE" echo "Generated: $(date)" >> "$OUTPUT_FILE"
for file in "$INPUT_DIR"/*; do echo "Processing $file..."

Add AI processing command here

done
undefined
INPUT_DIR="./incoming" OUTPUT_FILE="./outputs/report.md"
echo "# 批量处理报告" > "$OUTPUT_FILE" echo "生成时间：$(date)" >> "$OUTPUT_FILE"
for file in "$INPUT_DIR"/*; do echo "正在处理 $file..."

在此添加AI处理命令

done
undefined

2. Incremental Note Accumulation

2. 增量笔记累积

bash
#!/bin/bash
bash
#!/bin/bash

Append daily notes to a running log

将每日笔记追加到持续日志中

DATE=$(date +%Y-%m-%d) DAILY_LOG="./inbox/daily-notes-$DATE.md" WEEKLY_LOG="./outputs/weekly-log.md"
cat "$DAILY_LOG" >> "$WEEKLY_LOG" echo "" >> "$WEEKLY_LOG"
undefined
DATE=$(date +%Y-%m-%d) DAILY_LOG="./inbox/daily-notes-$DATE.md" WEEKLY_LOG="./outputs/weekly-log.md"
cat "$DAILY_LOG" >> "$WEEKLY_LOG" echo "" >> "$WEEKLY_LOG"
undefined

3. Template-Based Output Generation

3. 基于模板的输出生成

Always structure outputs with consistent headers:
markdown
undefined
始终使用一致的标题结构组织输出：
markdown
undefined

[Report Type]

[报告类型]

Generated: [Timestamp]
生成时间：[时间戳]

Summary

摘要

[High-level overview]
[高层概述]

Details

详情

[Structured findings]
[结构化发现]

Next Steps

后续步骤

[Action items]
Generated by OpenClaw Executive Assistant
undefined
[行动项]
由OpenClaw AI执行助理生成
undefined

Troubleshooting

故障排除

Issue: Prompts produce inconsistent output

问题：提示词输出不一致

Solution: Add explicit formatting instructions to prompts:
markdown
IMPORTANT: Use this exact structure:
解决方案： 在提示词中添加明确的格式说明：
markdown
重要提示：请严格使用以下结构：

Section Name

章节名称

  • Bullet point format only
  • No additional commentary
  • Maximum 3 sentences per item
undefined
  • 仅使用项目符号列表格式
  • 无额外评论
  • 每个条目最多3句话
undefined

Issue: File processing fails

问题：文件处理失败

Solution: Verify file permissions and paths:
bash
undefined
解决方案： 验证文件权限和路径：
bash
undefined

Check file permissions

检查文件权限

ls -la code-along/01-data-intake-review/incoming/
ls -la code-along/01-data-intake-review/incoming/

Fix permissions if needed

如有需要修复权限

chmod 644 code-along/01-data-intake-review/incoming/*
undefined
chmod 644 code-along/01-data-intake-review/incoming/*
undefined

Issue: Large email exports overwhelm processing

问题：大量邮件导出导致处理过载

Solution: Batch process in smaller groups:
bash
undefined
解决方案： 分成小批量处理：
bash
undefined

Process 10 emails at a time

每次处理10封邮件

ls eml/*.eml | head -10 | xargs -I {} cp {} eml/batch1/
ls eml/*.eml | head -10 | xargs -I {} cp {} eml/batch1/

Process batch1, then continue

处理batch1后继续

undefined
undefined

Best Practices

最佳实践

  1. Keep prompts version-controlled — Track prompt changes in git
  2. Review outputs before committing — AI-generated content should be verified
  3. Use consistent naming conventions — 
    YYYY-MM-DD-description.md
  4. Archive processed files — Move to 
    ./archive/
    after processing
  5. Maintain a changelog — Document workflow improvements
  1. 对提示词进行版本控制 — 在git中跟踪提示词的变更
  2. 提交前审核输出 — AI生成的内容需要验证
  3. 使用一致的命名规范 — 
    YYYY-MM-DD-描述.md
  4. 归档已处理文件 — 处理后移动到
    ./archive/
    文件夹
  5. 维护变更日志 — 记录工作流的改进

Extension Ideas

扩展思路

  • Mission Control Dashboard: Aggregate all reports into a single HTML view
  • Search Interface: Add full-text search across all generated reports
  • Automation Scripts: Create shell scripts to automate recurring workflows
  • Custom Prompt Library: Build organization-specific prompt templates
This workshop is designed for local-first, privacy-respecting AI workflows. All processing happens on your machine with your choice of AI assistant.
  • 任务控制仪表盘： 将所有报告整合到单个HTML视图中
  • 搜索界面： 添加对所有生成报告的全文搜索功能
  • 自动化脚本： 创建Shell脚本自动化重复工作流
  • 自定义提示词库： 构建组织专属的提示词模板
本工作坊专为本地优先、尊重隐私的AI工作流设计。所有处理均在你的设备上进行，可自由选择AI助理。