技能详情（站内镜像，无评论）
许可证：MIT-0
MIT-0 ·免费使用、修改和重新分发。无需归因。
版本：v1.0.1
统计：⭐ 0 · 66 · 1 current installs · 1 all-time installs
⭐ 0
安装量（当前） 1
🛡 VirusTotal ：良性 · OpenClaw ：可疑
Package：axelhu/health-check
安全扫描（ClawHub）
- VirusTotal ：良性
- OpenClaw ：可疑
OpenClaw 评估
The skill's stated purpose (daily health checks) matches the actions it describes, but the runtime instructions assume shell binaries, log access, and Feishu delivery credentials that are not declared — the manifest is incomplete and the instructions are vague about what logs and credentials will be used.
目的
The described checks (gateway status, disk, memory, recent logs, generate report) are coherent with a 'Health Check' skill. However the SKILL.md expects the 'openclaw' CLI and access to system logs and to a Feishu messaging channel, while the manifest declares no required binaries, env vars, or config paths — this is an inconsistency (the skill will require system tooling and messaging credentials to function).
说明范围
The runtime instructions instruct the agent to run shell commands (openclaw gateway status, df -h, free -h) and to 'check logs' but do not specify which log files or safe filters. They also mandate writing reports to data/exec-logs/... and sending the full report to Feishu. The combination lets the agent read arbitrary local logs and send their contents externally; the lack of explicit log paths and data filtering is a scope creep / data-exfil…
安装机制
This is instruction-only with no install steps or downloaded code, so nothing is written to disk by an installer. That minimizes install-time risks. The risk surface is runtime (shell commands and messaging).
证书
The instructions require sending messages to Feishu (a network endpoint) and likely need credentials or a webhook, but the skill declares no required environment variables or primary credential. That omission is problematic: either the agent must already have Feishu credentials elsewhere (not documented), or the skill will fail or implicitly rely on unspecified secrets. Also the skill reads system logs (sensitive data) yet provides no guidance…
持久
always is false and the skill does not request persistent presence or modify other skills' configuration. Writing reports to data/exec-logs is an expected behavior for a health-check tool, but users should confirm the directory is appropriate and writable.
安装（复制给龙虾 AI）
将下方整段复制到龙虾中文库对话中，由龙虾按 SKILL.md 完成安装。
请把本段交给龙虾中文库（龙虾 AI）执行：为本机安装 OpenClaw 技能「Health Check」。简介：每日安全检查。检查 OpenClaw Gateway、磁盘空间、内存使用等系统健康状态。触发时机：cron 定时任务或手动调用。。
请 fetch 以下地址读取 SKILL.md 并按文档完成安装：https://raw.githubusercontent.com/openclaw/skills/refs/heads/main/skills/axelhu/health-check/SKILL.md
（来源：yingzhi8.cn 技能库）SKILL.md
---
name: health-check
version: 1.0.0
description: 每日安全检查。检查 OpenClaw Gateway、磁盘空间、内存使用等系统健康状态。触发时机：cron 定时任务或手动调用。
---
# Health Check
每日系统安全检查，确保 OpenClaw 和环境正常运行。
## 核心流程
1. 读取 references/spec.md 获取详细规范
2. 检查 Gateway 运行状态
3. 检查磁盘空间
4. 检查内存使用
5. 检查最近日志有无错误
6. 生成报告并发送
## 触发时机
- cron 定时任务（建议每日）
- 用户明确要求时
## 投递规则（必须）
完成报告后，通过消息工具发送。
- 渠道：feishu
- 目标：<飞书群ID或用户ID>
如果报告超过 3800 字符，分成多条消息发送。
每条消息必须语义完整（不断开 URL 或格式）。
如果发送失败，重试一次。如果仍然失败，输出错误——永远不要静默退出。
## 输出
- 报告位置：`data/exec-logs/health-check/YYYY-MM-DD.md`
- 消息推送到飞书群