← 返回 Skills 市场
Multi Site Health Monitor
作者
ncreighton
· GitHub ↗
· v1.0.0
· MIT-0
247
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install multi-site-health-monitor
功能描述
Monitor dozens of websites with configurable health checks, auto-restart alerts, and intelligent alert routing. Use when the user needs uptime tracking, perf...
安全使用建议
This skill advertises many external integrations and actions (cloud restarts, SSH commands, Google Sheets, SMS, issue creation), but its declared requirements only include curl, jq and three API keys — several needed credentials and binaries are missing from the metadata. Before installing: (1) Ask the publisher to explain which integrations are actually implemented and to update requires.env and required binaries to list all needed credentials/tools (e.g., AWS keys or aws CLI, Google service account, SSH key, Twilio/Jira/GitHub tokens). (2) Request concrete runtime examples showing how credentials are used (scoped & least-privilege tokens). (3) If you test it, do so in a staging environment with limited, short-lived API keys and restricted PagerDuty/Datadog scopes; do not provide full admin credentials. (4) Consider restricting autonomous invocation (or disallow it) until the skill's actions and auth requirements are verified. If the publisher cannot justify the missing credentials/tools or refuses to provide code/rationale, treat the skill as untrusted and do not install it into production.
功能分析
Type: OpenClaw Skill
Name: multi-site-health-monitor
Version: 1.0.0
The skill bundle (SKILL.md) requests an extensive list of highly sensitive credentials, including AWS_SECRET_ACCESS_KEY, SSH_PRIVATE_KEY, and various API keys for PagerDuty, Datadog, and Slack. It instructs the AI agent to perform high-risk operations such as remote shell execution via SSH and infrastructure manipulation (AWS EC2 restarts) under the guise of a 'Multi-Site Health Monitor.' While these capabilities are plausibly related to automated incident response, the combination of extreme privilege requirements, the lack of actual implementation code (relying on the agent to use curl/jq), and the 'empire-skills' naming convention (potentially referencing post-exploitation frameworks) presents a significant security risk.
能力评估
Purpose & Capability
The README-style SKILL.md advertises wide-ranging integrations and privileged actions (AWS/Azure auto-restarts, executing shell commands via SSH, Google Sheets, WordPress monitoring, SMS notifications, creating Jira/GitHub issues). The declared requirements only list curl, jq and three env vars (SLACK_WEBHOOK_URL, PAGERDUTY_API_KEY, DATADOG_API_KEY). Required credentials and tools for AWS (aws CLI or AWS_ACCESS_KEY_ID/AWS_SECRET), Google APIs (OAuth client or service account), SSH private keys, Twilio/SMS credentials, WordPress admin credentials, or Jira/GitHub tokens are not declared — this is an incoherence between claimed purpose and requested capabilities.
Instruction Scope
SKILL.md contains operational instructions and examples that include POSTing to arbitrary restart webhooks, executing remote commands over SSH, triggering cloud actions, and writing to Google Sheets. Because this is an instruction-only skill with no code files, those instructions are what the agent will follow. They are broad and can cause side effects (remote restarts, posting to external endpoints). The document is vague about authentication and safe defaults, giving the agent significant latitude to call arbitrary endpoints and perform privileged actions — scope creep and potential for misuse.
Install Mechanism
No install spec and no code files — lowest install risk. Nothing is downloaded or written to disk by an installer as part of the skill bundle.
Credentials
Only three environment variables are required (Slack webhook, PagerDuty API key, Datadog API key), which make sense for some alerting features. However, the skill claims integrations that normally require additional credentials (AWS/Azure keys, Google service account, SSH keys, Twilio API creds, Jira/GitHub tokens). Those are not declared, so either the skill will not actually implement those integrations, or it will attempt to use other undeclared credentials or prompt for them at runtime. The declared env vars are thus insufficient for the advertised scope.
Persistence & Privilege
The skill is not marked always:true and does not request any system config paths. Autonomous invocation is allowed by platform default; that is expected. No evidence the skill attempts to modify other skills or agent-wide config.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install multi-site-health-monitor - 安装完成后，直接呼叫该 Skill 的名称或使用
/multi-site-health-monitor触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of Multi-Site Health Monitor, a production-grade uptime and incident monitoring skill.
- Monitor 10–100+ websites/services with flexible health checks (HTTP(s), TCP, DNS, SSL/TLS, Ping/ICMP).
- Intelligent alert routing to Slack, PagerDuty, Datadog, Google Sheets, and more with configurable thresholds and escalation rules.
- Automatic incident response: webhook triggers, auto-restart, AWS integration, and ticket creation for detected failures.
- Performance tracking: uptime calculations, error rates, historical reports, Datadog integration, compliance & audit logging.
- WordPress-specific monitoring: plugin/theme vulnerability checks, update/status alerts, backup and user activity verification.
- Requires environment setup for Slack, PagerDuty, Datadog (and optionally AWS, SSH, Google Sheets).
元数据
常见问题
Multi Site Health Monitor 是什么？
Monitor dozens of websites with configurable health checks, auto-restart alerts, and intelligent alert routing. Use when the user needs uptime tracking, perf... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 247 次。
如何安装 Multi Site Health Monitor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install multi-site-health-monitor」即可一键安装，无需额外配置。
Multi Site Health Monitor 是免费的吗？
是的，Multi Site Health Monitor 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Multi Site Health Monitor 支持哪些平台？
Multi Site Health Monitor 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（macos, linux, win32）。
谁开发了 Multi Site Health Monitor？
由 ncreighton（@ncreighton）开发并维护，当前版本 v1.0.0。
推荐 Skills