Automation Scripts

Manage automation scripts with creation, scheduling, logging, failure retries, and status notifications for monitoring, backup, sync, reports, and more.

Security Scan

The instructions generally match an automation-scripts purpose, but there are multiple mismatches and vagueness (references to a CLI that is not provided, platform-specific paths/services, and unspecified notification endpoints/credentials) that could hide risky behavior or make the skill non-functional.

Purpose & Capability

The SKILL.md describes creating/scheduling/logging automation scripts which is consistent with the name. However it repeatedly references a CLI command 'skill:automation-scripts' (no binary or install is declared) and platform-specific items like ~/.openclaw, 'Gateway进程', and 'OpenCode' integration. Those platform-specific accesses are not declared or justified in the metadata and make the declared purpose only partially coherent.

Instruction Scope

The instructions tell the agent/user to create and run shell scripts that check services, read configs, perform backups, run git auto-pushes, and call 'OpenCode' diagnostics. They also use vague placeholders like '发送到指定位置' without specifying destinations. This grants broad discretion and implies access to local system state and external endpoints beyond what is documented.

Install Mechanism

There is no install spec (instruction-only), which is lower risk. However, the README assumes a 'skill:automation-scripts' CLI exists; the absence of any declared binary or install steps is an incoherence — the skill as described cannot be invoked as shown unless extra software is present.

Credentials

No environment variables or credentials are requested, yet the instructions imply needing access to: user home files (~/.openclaw), service status (may require elevated rights), git remotes (would require credentials), and external services (OpenCode, notification endpoints). The absence of declared credentials is disproportionate to the operations the scripts describe.

Persistence & Privilege

The skill does not request 'always: true' and is user-invocable, which is appropriate. It does recommend creating scheduled jobs (cron) and writing files under ~/scripts and config locations; that is persistent activity but limited to the skill's own artifacts. Still, scheduling tasks that run arbitrary scripts increases risk if the scripts are later modified or sourced from untrusted places.

Scan Findings in Context

[no-findings] expected: Regex scanner found no code files to analyze — this is expected for an instruction-only skill. Absence of findings does not mean the instructions are safe or coherent.

What to consider before installing

Do not install or enable this skill without further details. Ask the publisher for: (1) the actual CLI/binary or installation steps referenced by 'skill:automation-scripts', (2) exact destinations for notifications and any external services (OpenCode) it will call, (3) what credentials (git, SSH, service tokens) are required and why, and (4) sample concrete scripts to review. If you must try it, run in an isolated/sandboxed environment, review every generated script and cron job before enabling, avoid supplying broad credentials, and verify backups are stored encrypted and not sent to unknown endpoints.

Automation Scripts Skill

自动化脚本管理 - 创建、执行、维护自动化任务

功能

• 脚本模板库
• 定时任务管理
• 执行日志记录
• 失败自动重试
• 状态通知

脚本分类

类别	说明	示例
监控	系统监控	健康检查、磁盘监控
备份	数据备份	配置备份、日志清理
同步	数据同步	Git自动提交、文件同步
报告	报告生成	日报、周报、月报
研究	自动化研究	GitHub项目分析

使用方法

# 列出可用脚本
skill:automation-scripts --list

# 创建新脚本
skill:automation-scripts --create --name "my-script" --type "monitor"

# 执行脚本
skill:automation-scripts --run "script-name"

# 查看执行日志
skill:automation-scripts --log "script-name"

# 定时执行
skill:automation-scripts --schedule "script-name" --cron "0 6 * * *"

# 禁用脚本
skill:automation-scripts --disable "script-name"

# 启用脚本
skill:automation-scripts --enable "script-name"

内置脚本模板

1. 健康检查脚本

#!/bin/bash
# OpenClaw 健康检查
# 位置: ~/scripts/openclaw-health-check.sh

# 检查项目
- 服务状态 (Gateway进程)
- 配置文件语法
- 磁盘空间
- 内存使用
- 日志错误

# 输出格式
✓ 检查通过
✗ 检查失败 (显示原因)

2. 自动备份脚本

#!/bin/bash
# 配置自动备份
# 位置: ~/scripts/backup-openclaw-config.sh

# 功能
- 每日自动备份配置
- 保留30天历史
- 清理旧备份

# 配置
BACKUP_DIR="~/.openclaw/backups"
RETENTION_DAYS=30

3. Git 自动同步脚本

#!/bin/bash
# Git 自动同步
# 位置: ~/scripts/auto-git-sync.sh

# 功能
- 检查文件变更
- 自动 add + commit
- 自动 push
- 记录同步日志

4. 定时报告脚本

#!/bin/bash
# 定时报告生成
# 位置: ~/scripts/daily-report.sh

# 功能
- 生成日报
- 发送到指定位置
- 记录发送状态

5. OpenCode 集成脚本

#!/bin/bash
# OpenCode 自动修复
# 位置: ~/scripts/opencode-auto-fix.sh

# 功能
- 检测服务故障
- 调用 OpenCode 诊断
- 自动修复
- 验证结果

脚本结构

scripts/
├── templates/           # 脚本模板
│   ├── monitor.sh
│   ├── backup.sh
│   ├── sync.sh
│   └── report.sh
├── custom/             # 自定义脚本
├── logs/               # 执行日志
└── config.conf         # 全局配置

配置示例

{
  "automation": {
    "enabled": true,
    "logRetentionDays": 30,
    "maxRetries": 3,
    "retryDelay": 60,
    "notifications": {
      "onFailure": true,
      "onSuccess": false
    }
  }
}

执行日志

字段	说明
timestamp	执行时间
script	脚本名称
status	success/failure
duration	执行耗时
output	输出摘要
error	错误信息（如有）

最佳实践

1. 脚本命名

   • 使用描述性名称
   • 包含类型前缀: backup-, monitor-, sync-

2. 错误处理

   • 总是检查返回值
   • 失败时发送通知
   • 保留错误日志

3. 日志管理

   • 记录执行详情
   • 定期清理旧日志
   • 分析失败模式

4. 定时任务

   • 避免高峰期执行
   • 设置合理重试
   • 监控执行状态