Agent-Skills.md

Agent Skills: Batch User Notification

Send IM messages to users in batch. Used for notifying specific user groups, sending after table filtering, all-staff notifications, etc. Use this Skill when administrators request batch notifications, mass messaging, or notifications after table filtering. Trigger words: notify/send/mass + users/batch/table.

UncategorizedID: Harryoung/efka/batch-notification

Repository

Harryoung/efka
HarryoungLicense: Apache-2.0
9117

Install this agent skill to your local

pnpm dlx add-skill https://github.com/Harryoung/efka/tree/HEAD/skills/batch-notification

Skill Files

Browse the full folder contents for batch-notification.

Download Skill

Loading file tree…

skills/batch-notification/SKILL.md

Skill Metadata

Name
batch-notification
Description
Send IM messages to users in batch. Used for notifying specific user groups, sending after table filtering, all-staff notifications, etc. Use this Skill when administrators request batch notifications, mass messaging, or notifications after table filtering. Trigger words: notify/send/mass + users/batch/table.

Batch User Notification

Support administrators to send IM notification messages to users in batch.

Typical Scenarios

  1. Upload table + filter conditions: Notify all users with benefits points greater than 0
  2. Upload target list: Notify specified user list
  3. All-staff notification: Notify everyone

Quick Start

All-staff Notification

mcp__{channel}__send_markdown_message(
    touser="@all",
    content="## Notification Title\n\nNotification content..."
)

Filtered Notification

python3 -c "
import pandas as pd
mapping = pd.read_excel('knowledge_base/企业管理/人力资源/user_mapping.xlsx')
business = pd.read_excel('/tmp/data.xlsx')
filtered = business[business['积分'] > 0]
result = pd.merge(filtered, mapping, on='工号', how='inner')
print('|'.join(result['企业微信用户ID'].tolist()))
"

Detailed Workflow

Complete 5-stage workflow, see WORKFLOW.md

pandas Query Patterns

Common filtering, JOIN, date processing patterns, see PANDAS_PATTERNS.md

Example Scenarios

Complete end-to-end examples, see EXAMPLES.md

Core Principles

  1. Privacy protection: Notifications are one-on-one private chats, messages must not contain other people's information
  2. Must confirm: Must wait for administrator reply "confirm send" after constructing message
  3. Python first: All table processing uses pandas
  4. Result transparency: Clearly report sending results (success/failure counts)

Available Tools

  • Bash: Execute pandas scripts
  • mcp__{channel}__send_markdown_message: Send Markdown messages
  • mcp__{channel}__send_text_message: Send plain text messages