Automation Scripts
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill is coherently about automation, but it describes unattended scripts that can automatically push code and repair services without clear approval, scoping, or rollback controls.
Use this skill only with explicit review before any script is run or scheduled. Review generated scripts, restrict them to known paths and repositories, require approval before commits/pushes or service fixes, and check where logs, backups, and scheduled jobs are stored.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If used as written, an automated script could publish repository changes or modify service behavior without the user reviewing each change.
The skill documents templates that automatically mutate Git repositories and automatically repair services, but the artifact does not provide approval, diff review, scope limits, sandboxing, or rollback requirements.
Git 自动同步 ... 自动 add + commit ... 自动 push; OpenCode 自动修复 ... 自动修复
Require explicit user approval before committing, pushing, or applying fixes; limit scripts to named repositories/paths; add dry-run, diff review, and rollback steps.
A mistaken script or bad diagnostic could be retried and spread across files, repositories, or services before the user notices.
Automatic retries combined with automatic pushes or repairs could repeatedly propagate a bad change into a repository or service, and the artifact does not describe containment beyond a retry count.
失败自动重试 ... "maxRetries": 3 ... 自动 push ... 自动修复
Use conservative retry policies, stop-on-mutation failures, alerts before retrying destructive actions, and clear rollback or restore procedures.
Scheduled scripts may continue running after the initial task and can keep affecting the local environment.
The skill explicitly supports scheduled recurring automation. This is central to its purpose and includes a disable command, but it creates persistent background behavior the user must manage.
skill:automation-scripts --schedule "script-name" --cron "0 6 * * *"; skill:automation-scripts --disable "script-name"
Keep an inventory of scheduled jobs, make disable/remove operations easy, and confirm schedules with the user before enabling them.
Local logs or backups could preserve sensitive information from scripts longer than expected.
The skill proposes retaining logs and configuration backups for 30 days. This is purpose-aligned, but logs and backups can contain sensitive command output or configuration details.
"logRetentionDays": 30; BACKUP_DIR="~/.openclaw/backups"; output | 输出摘要; error | 错误信息
Store logs and backups with restricted permissions, redact secrets from outputs, and let users configure retention and cleanup.