SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillslark-workflow-task-digest

lark-workflow-task-digest

Office & productivity

View repository

Updated 10-16 06:23

About this skill

Task Status Announcer Workflow

CRITICAL — Before starting, you MUST first use the Read tool to read ../lark-shared/SKILL.md, which contains authentication and permission handling

Applicable Scenarios

  • "How is today's task completion going" / "task progress"
  • "Generate this week's task weekly report" / "task report"
  • "See what to-dos I have" / "to-do report"
  • "project progress report" / "task overview"
  • "Help me announce the task status"

Prerequisites

Only supports user identity. Before execution, ensure authorization has been granted:

lark-cli auth login --domain task
# 如需推送消息或保存文档，还需：
lark-cli auth login --domain task,im,docs

Workflow

{时间范围 / 状态筛选}
  │
  ├── task +get-my-tasks [--page-all] ──► 任务列表
  │
  ├── AI 分类统计 ─────────────────────► 按状态/优先级/截止日分组
  │     ├── 未开始
  │     ├── 进行中
  │     ├── 已完成
  │     └── 已逾期（高亮）
  │
  ├── AI 生成进度报告 ─────────────────► 结构化 Markdown
  │
  ├── [可选] im +messages-send ────────► 推送到飞书
  │
  └── [可选] docs +create ─────────────► 保存为文档

Step 1: Determine the time range

Determine the query range based on the user's request:

User says Range
Today / today The current day
This week This Monday ~ today
This month The 1st of this month ~ today

Note: Use the system command for date calculation date, do not calculate mentally.

# 获取今天的截止范围（示例）
date -d "today 23:59:59" +%s   # Unix timestamp

Step 2: Get the task list

# 默认：返回分配给当前用户的未完成任务（最多 20 条）
lark-cli task +get-my-tasks

# 获取全部未完成任务（超过 20 条时）
lark-cli task +get-my-tasks --page-all

# 按截止日期筛选（推荐，减少数据量）
lark-cli task +get-my-tasks --due-end "2026-04-09T23:59:59+08:00"

# 按创建时间筛选
lark-cli task +get-my-tasks --created-at "2026-04-01"

# 包含已完成的任务
lark-cli task +get-my-tasks --complete

# 搜索特定任务
lark-cli task +get-my-tasks --query "<关键词>"

Note: Without filter conditions, a large number of historical to-dos may be returned; it is recommended to use --due-end to filter. If the data volume is still too large (exceeding the context limit), when summarizing, the AI should only displaythose created within the last 30 days, and collapse the rest as "N other historical to-dos".

Step 3: AI generates the progress report

Organize the results of Step 2 according to the following structure:

## {日期范围} 任务进度报告

### 统计概览
| 状态 | 数量 |
|------|------|
| 未开始 | N |
| 进行中 | N |
| 已完成 | N |
| 已逾期 | N |
| **总计** | **N** |

### 逾期任务（需关注）
| 任务 | 负责人 | 截止日期 | 逾期天数 |
|------|--------|---------|---------|

### 进行中任务
| 任务 | 负责人 | 截止日期 | 进度 |
|------|--------|---------|------|

### 已完成任务
| 任务 | 完成时间 |
|------|---------|

Statistical rules:

  • Overdue = due date < current time and status != completed
  • Completion rate = completed / total × 100%
  • Overdue taskshighlighted(marked in red)

Step 4: Push the report (optional)

Send to a Feishu group

lark-cli im +messages-send \
  --chat-id "<chat_id>" \
  --markdown "{报告内容}" \
  --as bot

Send to an individual

lark-cli im +messages-send \
  --user-id "<open_id>" \
  --markdown "{报告内容}" \
  --as bot

Save as a document

lark-cli docs +create \
  --title "任务进度报告 ({日期})" \
  --markdown "{报告内容}" \
  --as user

Fallback Strategy

Scenario Fallback Plan
No task data Return "There are currently no tasks matching the criteria"
Data volume too large Only display data from the last 30 days, collapse historical data
im push failed Only output to the conversation, inform the user to send manually
docs save failed Only output Markdown to the conversation

Permission Table

Command Required scope
task +get-my-tasks task:task:readonly
im +messages-send im:message
docs +create docx:document:create

References