Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问clear审计未展示

notification-agent通知 Agent

Agent Skill

notification-agent 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

729

周安装

31

GitHub Stars

公开资料未说明

下载量

255
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:notification-agent(通知 Agent)
来源仓库:https://github.com/psh355q-ui/szdi57465yt
仓库路径:skills/notification-agent
安装命令:
npx skills add psh355q-ui/szdi57465yt --skill "notification-agent"
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

AgentSkills.tonpx skills
npx skills add psh355q-ui/szdi57465yt --skill "notification-agent"

简介

notification-agent 用于查找、检索和筛选相关信息,辅助任务调度与状态跟踪。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据线索快速定位候选结果时使用。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装并启用该技能。
  • 需检查是否依赖外部 API 或数据库,确认网络权限与数据安全措施到位。
  • 建议在沙箱环境中先行测试,确保不影响主业务流程稳定性。

SKILL.md

name
notification-agent
description
Multi-channel notification dispatcher. Sends trading alerts, reports, and system notifications via Telegram, Slack, Email, and WebSocket. Supports urgency-based routing and custom formatting.
license
Proprietary
compatibility
Requires Telegram Bot API, Slack API, SMTP, WebSocket
metadata
author
ai-trading-system
version
1.0
category
system
agent_role
notifier

Notification Agent - 알림 발송 관리자

Role

Trading Signal, 리포트, 시스템 알림을 Telegram, Slack, Email, WebSocket을 통해 적절한 채널로 전송합니다.

Core Capabilities

1. Multi-Channel Support

Telegram

  • Use Case: 실시간 거래 Signal, 긴급 알림
  • Format: Markdown with buttons
  • Priority: High urgency

Slack

  • Use Case: 팀 협업, 일일 리포트
  • Format: Rich formatting with attachments
  • Priority: Medium urgency

Email

  • Use Case: 주간/월간 리포트
  • Format: HTML with charts
  • Priority: Low urgency

WebSocket

  • Use Case: Dashboard 실시간 업데이트
  • Format: JSON
  • Priority: Real-time

2. Urgency-Based Routing

URGENCY_ROUTING = {
    'CRITICAL': ['Telegram', 'Slack', 'WebSocket'],  # 즉시 모든 채널
    'HIGH': ['Telegram', 'WebSocket'],                # 즉시 알림
    'MEDIUM': ['Slack', 'Email'],                     # 배치 발송
    'LOW': ['Email']                                  # 일일 요약에 포함
}

3. Message Templates

Trading Signal Template (Telegram)

🎯 **New Trading Signal**

**Ticker**: {ticker}
**Action**: {action}
**Confidence**: {confidence:.0%}
**Source**: {source}

**Reasoning**: {reasoning}

**Target**: ${target_price}
**Stop Loss**: ${stop_loss}

[Approve] [Reject]

Daily Report Template (Email)

<h1>Daily Trading Report - {date}</h1>

<h2>Performance</h2>
<table>
  <tr><td>Win Rate</td><td>{win_rate:.1%}</td></tr>
  <tr><td>Daily Return</td><td>{return:.2%}</td></tr>
</table>

<h2>Top Signals</h2>
...

4. Rate Limiting

RATE_LIMITS = {
    'Telegram': 30 / 60,      # 30 messages per minute
    'Slack': 1 / 1,            # 1 message per second
    'Email': 100 / 3600,       # 100 emails per hour
    'WebSocket': None          # No limit
}

Decision Framework

Step 1: Receive Notification Request
  - Type: signal, report, alert, error
  - Urgency: critical, high, medium, low
  - Content: message body
  - Recipients: list of users/channels

Step 2: Determine Channels
  Based on urgency:
    CRITICAL → All channels
    HIGH → Telegram + WebSocket
    MEDIUM → Slack + Email
    LOW → Email only

Step 3: Format Message
  For each channel:
    - Apply channel-specific template
    - Format content (Markdown, HTML, JSON)
    - Add buttons/actions if applicable

Step 4: Check Rate Limits
  IF rate limit exceeded:
    → Queue message
    → Send when available

Step 5: Send Notification
  Try:
    send_to_channel(channel, formatted_message)
  Except:
    log_error()
    retry_with_backoff()

Step 6: Track Delivery
  - Log sent time
  - Track delivery status
  - Record user interaction (if applicable)

Output Format

{
  "notification_id": "NOTIF-20251221-001",
  "type": "trading_signal",
  "urgency": "HIGH",
  "content": {
    "ticker": "AAPL",
    "action": "BUY",
    "confidence": 0.85,
    "source": "war_room",
    "reasoning": "Strong consensus...",
    "target_price": 205.00,
    "stop_loss": 195.00
  },
  "channels": ["telegram", "websocket"],
  "recipients": {
    "telegram": ["COMMANDER_CHAT_ID"],
    "websocket": ["active_connections"]
  },
  "sent_at": "2025-12-21T13:00:00Z",
  "delivery_status": {
    "telegram": {
      "status": "sent",
      "message_id": "12345",
      "sent_at": "2025-12-21T13:00:01Z"
    },
    "websocket": {
      "status": "broadcasted",
      "connections": 3,
      "sent_at": "2025-12-21T13:00:00Z"
    }
  }
}

Examples

Example 1: 긴급 Trading Signal (CRITICAL)

Input:
- Type: trading_signal
- Urgency: CRITICAL
- Content: Emergency FDA approval for MRNA

Channels:
- Telegram: Immediate alert with [Approve] button
- Slack: Rich message with details
- WebSocket: Real-time dashboard update

Output:
- All channels notified within 5 seconds

Example 2: 일일 리포트 (MEDIUM)

Input:
- Type: daily_report
- Urgency: MEDIUM
- Content: Daily performance summary

Channels:
- Slack: Summary card
- Email: Full HTML report

Output:
- Slack: Posted to #trading channel
- Email: Sent to  [email protected] 

Example 3: Circuit Breaker 발동 (CRITICAL)

Input:
- Type: emergency_alert
- Urgency: CRITICAL
- Content: Circuit Breaker triggered (Daily Loss > -2%)

Channels:
- Telegram: URGENT alert
- Slack: @channel mention
- Email: High priority

Output:
- Immediate notification to all channels
- Telegram bot calls Commander

Example 4: WebSocket 실시간 업데이트 (HIGH)

Input:
- Type: new_signal
- Urgency: HIGH
- Content: New signal from Deep Reasoning

Channels:
- WebSocket: Broadcast to /trading page

Output:
- Dashboard updates instantly
- No Telegram/Email (not urgent enough for push)

Guidelines

Do's ✅

  • 적절한 채널 선택: 긴급도에 맞게
  • Rate Limit 준수: 스팸 방지
  • Clear Formatting: 읽기 쉽게
  • Action Buttons: 즉시 조치 가능하게

Don'ts ❌

  • 과도한 알림 금지 (Notification fatigue)
  • 중요하지 않은 것을 CRITICAL로 표시 금지
  • 에러 메시지를 사용자에게 직접 노출 금지
  • Rate limit 초과 금지

Integration

Telegram Bot

from backend.notifications.telegram_commander_bot import TelegramCommanderBot

telegram = TelegramCommanderBot(
    bot_token=os.getenv('TELEGRAM_BOT_TOKEN'),
    commander_chat_id=os.getenv('TELEGRAM_COMMANDER_CHAT_ID')
)

async def send_telegram_signal(signal: Dict):
    """Send trading signal via Telegram"""
    
    message = f"""
🎯 **New Trading Signal**

**Ticker**: {signal['ticker']}
**Action**: {signal['action']}
**Confidence**: {signal['confidence']:.0%}

**Reasoning**: {signal['reasoning']}

**Target**: ${signal['target_price']}
**Stop Loss**: ${signal['stop_loss']}
"""
    
    # Add approval buttons
    keyboard = {
        "inline_keyboard": [[
            {"text": "✅ Approve", "callback_data": f"approve_{signal['signal_id']}"},
            {"text": "❌ Reject", "callback_data": f"reject_{signal['signal_id']}"}
        ]]
    }
    
    await telegram.send_message(
        text=message,
        parse_mode='Markdown',
        reply_markup=keyboard
    )

WebSocket Broadcast

from fastapi import WebSocket

active_connections: List[WebSocket] = []

async def broadcast_signal(signal: Dict):
    """Broadcast signal to all connected clients"""
    
    message = {
        "type": "new_signal",
        "data": signal
    }
    
    for connection in active_connections:
        try:
            await connection.send_json(message)
        except:
            active_connections.remove(connection)

Email Report

import smtplib
from email.mime.text import MIMEText
from email.mime.multipart import MIMEMultipart

def send_email_report(report_html: str, recipient: str):
    """Send HTML email report"""
    
    msg = MIMEMultipart('alternative')
    msg['Subject'] = f"Daily Trading Report - {datetime.now().strftime('%Y-%m-%d')}"
    msg['From'] = os.getenv('SMTP_FROM')
    msg['To'] = recipient
    
    html_part = MIMEText(report_html, 'html')
    msg.attach(html_part)
    
    with smtplib.SMTP(os.getenv('SMTP_HOST'), int(os.getenv('SMTP_PORT'))) as server:
        server.starttls()
        server.login(os.getenv('SMTP_USER'), os.getenv('SMTP_PASSWORD'))
        server.send_message(msg)

Slack Integration

from slack_sdk.webhook import WebhookClient

slack = WebhookClient(os.getenv('SLACK_WEBHOOK_URL'))

def send_slack_report(report: Dict):
    """Send report to Slack"""
    
    blocks = [
        {
            "type": "header",
            "text": {
                "type": "plain_text",
                "text": f"📊 Daily Report - {report['date']}"
            }
        },
        {
            "type": "section",
            "fields": [
                {"type": "mrkdwn", "text": f"*Win Rate:*\n{report['win_rate']:.1%}"},
                {"type": "mrkdwn", "text": f"*Return:*\n{report['return']:.2%}"}
            ]
        }
    ]
    
    slack.send(blocks=blocks)

Rate Limiting Implementation

from collections import deque
from time import time

class RateLimiter:
    def __init__(self, max_calls: int, period: float):
        self.max_calls = max_calls
        self.period = period
        self.calls = deque()
    
    def allow(self) -> bool:
        """Check if call is allowed"""
        now = time()
        
        # Remove old calls
        while self.calls and self.calls[0] < now - self.period:
            self.calls.popleft()
        
        # Check limit
        if len(self.calls) < self.max_calls:
            self.calls.append(now)
            return True
        
        return False

# Usage
telegram_limiter = RateLimiter(max_calls=30, period=60)

if telegram_limiter.allow():
    await send_telegram_message(msg)
else:
    queue_message(msg)  # Send later

Performance Metrics

  • Delivery Success Rate: > 99%
  • Latency (CRITICAL): < 5 seconds
  • Latency (HIGH): < 30 seconds
  • Rate Limit Violations: 0

Notification Queue

from queue import PriorityQueue

notification_queue = PriorityQueue()

# Priority: CRITICAL=1, HIGH=2, MEDIUM=3, LOW=4
notification_queue.put((1, critical_notification))
notification_queue.put((3, medium_notification))

# Worker processes queue
while True:
    priority, notification = notification_queue.get()
    send_notification(notification)

Version History

  • v1.0 (2025-12-21): Initial release with Telegram, Slack, Email, WebSocket support

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

04

需要参考平台分布和安装热度时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Claude Code

26.21%
按下载量换算67

Antigravity

23.35%
按下载量换算60

windsurf

16.26%
按下载量换算41

trae

12.75%
按下载量换算33

OpenCode

8%
按下载量换算20

Gemini CLI

3.46%
按下载量换算9

安全审计

暂无安全审计结果可展示。

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills