tasq/node_modules/@claude-flow/cli/.claude/commands/agents/health.md

name	description	type
health	Show agent health and metrics	command

Agent Health Command

Monitor agent health status, resource usage, and detect issues.

Usage

npx claude-flow agent health [agent-id] [options]

Options

Option	Short	Description	Default
--watch	-w	Continuous monitoring	false
--interval	-i	Watch interval in seconds	5
--format		Output format (table, json)	table

Examples

# Overall health check
npx claude-flow agent health

# Specific agent health
npx claude-flow agent health coder-lx7m9k2

# Continuous monitoring
npx claude-flow agent health --watch

# Custom interval
npx claude-flow agent health -w -i 10

# JSON output
npx claude-flow agent health --format json

Output

Agent Health Status

Overall: HEALTHY

+--------------------+---------+--------+--------+---------+
| Agent              | Status  | Memory | CPU    | Tasks   |
+--------------------+---------+--------+--------+---------+
| coder-lx7m9k2      | healthy | 45MB   | 2.3%   | 5/5     |
| researcher-abc123  | healthy | 38MB   | 1.8%   | 3/3     |
| tester-def456      | warning | 120MB  | 8.5%   | 12/14   |
+--------------------+---------+--------+--------+---------+

Alerts
  - tester-def456: High memory usage (120MB > 100MB threshold)
  - tester-def456: Task failure rate above threshold (14.3%)

Recommendations
  - Consider restarting tester-def456
  - Review failed tasks for error patterns

Health Status Values

Status	Description
healthy	All metrics within normal range
warning	Some metrics approaching thresholds
critical	Metrics exceeded critical thresholds
unknown	Unable to determine health

Health Checks

Resource Checks

• Memory usage vs threshold
• CPU utilization
• Active connections
• Queue depth

Performance Checks

• Response time
• Task success rate
• Error frequency
• Throughput

Connectivity Checks

• MCP server connection
• Memory backend
• Neural services
• Swarm coordination

Watch Mode

Continuous monitoring with real-time updates:

npx claude-flow agent health --watch

# Output updates every 5 seconds:
# [10:30:15] Agent Health: 3 healthy, 0 warning, 0 critical
# [10:30:20] Agent Health: 3 healthy, 0 warning, 0 critical
# [10:30:25] Agent Health: 2 healthy, 1 warning, 0 critical
#   - tester-def456: Memory usage increased to 115MB

Press Ctrl+C to stop watching.

JSON Output

{
  "overall": "healthy",
  "agents": [
    {
      "id": "coder-lx7m9k2",
      "status": "healthy",
      "metrics": {
        "memoryMB": 45,
        "cpuPercent": 2.3,
        "tasksCompleted": 5,
        "tasksTotal": 5
      }
    }
  ],
  "alerts": [],
  "recommendations": []
}

Related Commands

• npx claude-flow agent status - Detailed agent info
• npx claude-flow agent metrics - Performance metrics
• npx claude-flow doctor - System-wide health