error-monitoring-agent（错误监控 Agent）

Error Monitoring Agent

Real-time error monitoring and alerting for AI agents. Detect, track, analyze, and resolve errors automatically.

Overview

A comprehensive error monitoring system that helps agents detect exceptions in real-time, track error patterns, set up intelligent alerts, and automate resolution workflows.

Capabilities

1. Error Detection

node monitor.js watch --source logs,api,workers --threshold 5/min
node monitor.js watch --pattern "UnhandledPromiseRejection|ENOTFOUND"

Monitors multiple sources for errors with configurable thresholds and pattern matching.

2. Error Aggregation

node monitor.js aggregate --group-by stacktrace --min-similarity 0.85
node monitor.js aggregate --time-window 1h --top 20

Groups similar errors together to reduce noise and identify patterns.

3. Alert Rules

node monitor.js alert --rule "error_rate > 10/min" --channel slack
node monitor.js alert --rule "new_error_type" --channel pagerduty --severity critical
node monitor.js alert --rule "error_spike > 3x_baseline" --channel email

Configurable alerting with rate thresholds, new error detection, and spike monitoring.

4. Root Cause Analysis

node monitor.js analyze --error-id err_abc123 --depth 5
node monitor.js analyze --correlate deploy-log,config-change

Traces error chains, correlates with deployments and config changes.

5. Auto-Resolution

node monitor.js auto-resolve --strategy restart,retry,rollback
node monitor.js auto-resolve --known-fixes db --apply-approved

Automatically resolves known error patterns with approved remediation strategies.

Configuration

{
  "monitoring": {
    "sources": ["application", "infrastructure", "api"],
    "sampling": 1.0,
    "retention": "30d",
    "alertRules": [
      { "condition": "error_rate > 10/min", "action": "page-oncall" },
      { "condition": "new_error_type", "action": "notify-channel" },
      { "condition": "error_spike > 3x", "action": "auto-investigate" }
    ],
    "autoResolve": {
      "enabled": true,
      "approvedStrategies": ["restart-service", "retry-request", "rollback-deploy"]
    }
  }
}

Use Cases

• Production Monitoring: Watch production systems for errors 24/7
• CI/CD Integration: Monitor deployment health after releases
• Agent Health: Track AI agent errors and failures
• Incident Response: Detect and respond to incidents automatically
• Error Budgets: Track error rates against SLO targets