← 返回 Skills 市场

Agent Health Monitor

Name: Agent Health Monitor
Author: jpengcheng523-netizen

作者 jpengcheng523-netizen · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

164

总下载

当前安装

版本数

在 OpenClaw 中安装

/install agent-health-monitor

功能描述

Monitors agent health status and detects failures for fault-tolerant agent systems. **Trigger scenarios:** - User asks about agent health or status - User me...

使用说明 (SKILL.md)

Agent Health Monitor

Monitors agent and session health to detect failures and ensure system reliability.

Usage

const monitor = require('./skills/agent-health-monitor');

// Check all agents health
const health = await monitor.checkHealth();

// Get failed agents
const failed = await monitor.getFailedAgents();

// Monitor continuously
monitor.startMonitoring(30000); // Check every 30s

Features

Session Status Check: Verifies active sessions
Agent Liveness Detection: Detects unresponsive agents
Resource Monitoring: Tracks CPU/memory usage
Failure Alerts: Notifies on detected failures

Output

Returns a health report with:

Total agents/sessions count
Healthy vs unhealthy count
Failed agents list
Resource usage summary

安全使用建议

This skill largely does what it claims, but it runs shell commands and reads files under your home directory: it calls the openclaw CLI, reads /proc and /proc/meminfo (Linux-specific), and checks ~/.openclaw and a hard-coded workspace path (feishu-evolver-wrapper) by running node lifecycle.js there. Before installing: (1) review the index.js source to confirm the wrapper path and commands are acceptable in your environment; (2) ensure any lifecycle.js or wrapper code in ~/.openclaw/workspace/skills is trusted (the monitor will invoke it); (3) be aware this runs local commands and inspects files but does not contact external endpoints or request credentials; (4) consider running it in a restricted/sandboxed environment first. If you need tighter assurance, ask the author to document all filesystem paths and shell commands used, or modify the code to make paths configurable rather than hard-coded.

功能分析

Type: OpenClaw Skill Name: agent-health-monitor Version: 1.0.0 The skill monitors system and agent health using high-risk capabilities, including shell command execution via `execSync` and direct filesystem access via `fs` in `index.js`. It performs broad checks on system resources (`/proc/meminfo`, `/proc/loadavg`) and specifically attempts to execute a lifecycle script from another skill's directory (`feishu-evolver-wrapper`), which is an unusual cross-skill interaction. While these actions are plausibly needed for the stated purpose, the use of powerful system APIs and the execution of external scripts from the workspace meet the threshold for a suspicious classification.

能力评估

ℹ Purpose & Capability

The code implements session listing, liveness checks, resource monitoring, and wrapper status checks which align with an agent health monitor. One unexpected hard-coded target is the '.openclaw/workspace/skills/feishu-evolver-wrapper' wrapper path — this ties the monitor to a specific wrapper component not mentioned in the README.

⚠ Instruction Scope

SKILL.md describes the monitoring API but does not enumerate the runtime actions the code performs: it executes local shell commands (openclaw CLI, cat /proc files, node lifecycle.js), reads files under ~/.openclaw, and inspects skill workspace contents. Those file and command accesses are broader than the metadata (which declared no config paths) and should be documented.

✓ Install Mechanism

No install spec or external downloads — the skill is a local Node module. No package installation from remote URLs occurs.

ℹ Credentials

The package declares no required environment variables or credentials, but the code reads process.env.HOME and accesses files under the user's home directory (~/.openclaw). While HOME is standard, the un-declared config path access (workspace skills directory) is a modest mismatch with metadata and should be documented.

✓ Persistence & Privilege

The skill does not request 'always: true' or modify other skills' configs. It runs on demand or via startMonitoring; autonomous invocation is allowed by default but not a special privilege here.

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 agent-health-monitor
安装完成后，直接呼叫该 Skill 的名称或使用 /agent-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 agent-health-monitor. - Monitors agent and session health to detect failures and maintain reliability. - Detects unresponsive agents and checks active sessions. - Tracks CPU and memory usage for resource monitoring. - Provides failure alerts and detailed health reports, including status counts and usage summaries.

元数据

Slug agent-health-monitor

版本 1.0.0

许可证 MIT-0

累计安装 1

当前安装数 1

历史版本数 1

常见问题