Starsoup Docs

Appearance

晨间报告 Cron 迁移计划 ​

For Claude: REQUIRED SUB-SKILL: Use superpowers:executing-plans to implement this plan task-by-task.

Goal: 將 OpenClaw cron 晨间报告遷移到獨立 Python 腳本 + Systemd Timer，消除每日 Kimi token 消耗

Architecture:

  • Python 腳本直接調用 API（Moonshot balance、Vikunja tasks、Docker status）
  • Discord Webhook 推送報告（無需 OpenClaw message 工具）
  • Systemd Timer 替代 OpenClaw cron，每天 10:00 JST 運行
  • 保留現有報告格式，僅改變執行方式

Tech Stack: Python 3.12+、Systemd、Discord Webhook、Vikunja API、Moonshot API

Related: 項目規則、Obsidian文檔規則

背景與問題 ​

當前 OpenClaw cron 任務 晨间报告 - 东京时间10点 使用 kimi-k2.5 模型，每次運行消耗約 15k-35k tokens：

  • 過去幾週多次運行記錄顯示 input tokens 持續累積
  • 高峰期單次運行達 169k input tokens
  • 對於單純的數據收集和格式化任務，這是不必要的 AI 資源消耗

解決方案： 將報告邏輯遷移到純 Python 腳本，通過 systemd timer 觸發，完全消除 AI token 消耗。

數據需求分析 ​

原始 cron prompt 要求收集以下數據：

數據類別來源API/命令
Moonshot API 餘額Moonshot APIGET /v1/users/me/balance
服務器運行時間Systemuptime
系統負載Systemcat /proc/loadavg
內存使用Systemfree -h 或 /proc/meminfo
磁盤空間Systemdf -h /
Docker 容器狀態Dockerdocker ps -a --format
Vikunja Doing 任務Vikunja APIGET /api/v1/projects/{id}/views/{vid}/tasks

Task 1: 創建 Python 腳本目錄結構 ​

Files:

  • Create: ~/.openclaw/workspace/projects/morning-report/ (項目目錄)
  • Create: ~/.openclaw/workspace/projects/morning-report/src/ (源碼目錄)
  • Create: ~/.openclaw/workspace/projects/morning-report/.env.example (環境變量模板)

Step 1: 創建目錄結構

bash
mkdir -p ~/.openclaw/workspace/projects/morning-report/src
cd ~/.openclaw/workspace/projects/morning-report
touch .env.example README.md

Step 2: Commit

bash
cd ~/.openclaw/workspace
git add projects/morning-report/
git commit -m "infra: create morning-report project structure"

Task 2: 實現配置管理模塊 ​

Files:

  • Create: ~/.openclaw/workspace/projects/morning-report/src/config.py
  • Create: ~/.openclaw/workspace/projects/morning-report/.env (gitignored)

Step 1: 創建配置模塊

python
"""Configuration management for morning report."""
import os
from dataclasses import dataclass
from pathlib import Path

@dataclass
class Config:
    # Moonshot API
    moonshot_api_key: str
    
    # Vikunja API
    vikunja_base_url: str = "http://localhost:8081"
    vikunja_token: str = ""
    
    # Discord Webhook
    discord_webhook_url: str = ""
    discord_channel_id: str = "1470638978716668061"  # #healthcheck
    
    # Report settings
    timezone: str = "Asia/Tokyo"
    
    @classmethod
    def from_env(cls) -> "Config":
        """Load configuration from environment variables."""
        return cls(
            moonshot_api_key=os.getenv("MOONSHOT_API_KEY", ""),
            vikunja_base_url=os.getenv("VIKUNJA_BASE_URL", "http://localhost:8081"),
            vikunja_token=os.getenv("VIKUNJA_TOKEN", ""),
            discord_webhook_url=os.getenv("DISCORD_WEBHOOK_URL", ""),
            discord_channel_id=os.getenv("DISCORD_CHANNEL_ID", "1470638978716668061"),
            timezone=os.getenv("TIMEZONE", "Asia/Tokyo"),
        )
    
    def validate(self) -> None:
        """Validate required configuration."""
        required = ["moonshot_api_key", "vikunja_token", "discord_webhook_url"]
        missing = [f for f in required if not getattr(self, f)]
        if missing:
            raise ValueError(f"Missing required config: {', '.join(missing)}")

Step 2: 創建 .env.example

bash
cat > ~/.openclaw/workspace/projects/morning-report/.env.example << 'EOF'
# Moonshot API
MOONSHOT_API_KEY=sk-...

# Vikunja API
VIKUNJA_BASE_URL=http://localhost:8081
VIKUNJA_TOKEN=tk_...

# Discord Webhook
DISCORD_WEBHOOK_URL=https://discord.com/api/webhooks/...
DISCORD_CHANNEL_ID=1470638978716668061

# Timezone
TIMEZONE=Asia/Tokyo
EOF

Step 3: 創建 .gitignore

bash
cat > ~/.openclaw/workspace/projects/morning-report/.gitignore << 'EOF'
.env
__pycache__/
*.pyc
*.pyo
*.pyd
.Python
*.so
*.egg
*.egg-info/
dist/
build/
.venv/
venv/
EOF

Step 4: Commit

bash
cd ~/.openclaw/workspace
git add projects/morning-report/
git commit -m "feat(morning-report): add config management"

後續任務摘要 ​

由於內容較長，以下為剩餘任務的概要：

Task 3: 實現數據收集模塊 (collectors.py) ​

  • SystemCollector: 收集 uptime、load avg、memory、disk
  • MoonshotCollector: 調用 Moonshot API 獲取餘額
  • DockerCollector: 調用 docker ps 獲取容器狀態
  • VikunjaCollector: 調用 Vikunja API 獲取 Doing 任務

Task 4: 實現報告生成模塊 (report.py) ​

  • ReportGenerator.generate(): 將數據格式化為 Markdown 報告
  • 保持與現有報告格式一致

Task 5: 實現 Discord 發送模塊 (notifier.py) ​

  • DiscordNotifier.send(): 通過 webhook 發送報告

Task 6: 實現主入口腳本 (main.py) ​

  • 整合所有模塊
  • 加載配置、收集數據、生成報告、發送通知

Task 7: 創建 Systemd Service 和 Timer ​

  • starsoup-morning-report.service: oneshot service
  • starsoup-morning-report.timer: 每天 10:00 JST 運行
  • install.sh: 自動安裝腳本

Task 8-12: 配置、測試、安裝、移除舊 Cron、文檔 ​

成本節省估算 ​

指標之前 (OpenClaw cron)之後 (Python 腳本)
每日 Token 消耗~20k-35k0
每日成本~¥0.10-0.20¥0
每月節省~¥3-6-