tapps-agents 3.5.41__py3-none-any.whl → 3.6.1__py3-none-any.whl

tapps_agents/cli/commands/health.py

@@ -1,665 +1,665 @@
-"""
-Health command handlers.
-"""
-
-from __future__ import annotations
-
-import json
-import logging
-import sys
-from collections import defaultdict
-from datetime import UTC, datetime, timedelta
-from pathlib import Path
-
-from ...health.checks.automation import AutomationHealthCheck
-from ...health.checks.environment import EnvironmentHealthCheck
-from ...health.checks.execution import ExecutionHealthCheck
-from ...health.checks.context7_cache import Context7CacheHealthCheck
-from ...health.checks.knowledge_base import KnowledgeBaseHealthCheck
-from ...health.checks.outcomes import OutcomeHealthCheck
-from ...health.collector import HealthMetricsCollector
-from ...health.dashboard import HealthDashboard
-from ...health.orchestrator import HealthOrchestrator
-from ...health.registry import HealthCheckRegistry
-from ..feedback import get_feedback, ProgressTracker
-from .common import format_json_output
-
-
-def _usage_data_from_execution_metrics(project_root: Path) -> dict | None:
-    """
-    Build usage-like data from execution metrics when analytics is empty.
-
-    Aggregates .tapps-agents/metrics/executions_*.jsonl by today (steps/workflows),
-    by skill (agents), and by workflow_id (workflows). Returns same shape as
-    AnalyticsDashboard.get_dashboard_data() for system/agents/workflows.
-    """
-    try:
-        from ...workflow.execution_metrics import ExecutionMetricsCollector
-
-        collector = ExecutionMetricsCollector(project_root=project_root)
-        metrics = collector.get_metrics(limit=5000)
-        if not metrics:
-            return None
-
-        now = datetime.now(UTC)
-        today_start = now.replace(hour=0, minute=0, second=0, microsecond=0)
-        thirty_days_ago = now - timedelta(days=30)
-
-        # Filter to last 30 days
-        def parse_ts(ts: str) -> datetime:
-            return datetime.fromisoformat(ts.replace("Z", "+00:00"))
-
-        recent = [m for m in metrics if parse_ts(m.started_at) >= thirty_days_ago]
-        if not recent:
-            return None
-
-        today_metrics = [m for m in recent if parse_ts(m.started_at) >= today_start]
-        workflow_ids_today_success = {
-            m.workflow_id for m in today_metrics if m.status == "success"
-        }
-        workflow_ids_today_failed = {
-            m.workflow_id for m in today_metrics if m.status != "success"
-        }
-        completed_today = len(workflow_ids_today_success)
-        failed_today = len(workflow_ids_today_failed)
-        avg_duration = (
-            sum(m.duration_ms for m in recent) / len(recent) / 1000.0
-            if recent
-            else 0.0
-        )
-
-        # Agents: by skill or command
-        agent_counts: dict[str, list] = defaultdict(list)
-        for m in recent:
-            key = m.skill or m.command or "unknown"
-            agent_counts[key].append(m)
-
-        agents_list = []
-        for name, ms in agent_counts.items():
-            total = len(ms)
-            success = sum(1 for m in ms if m.status == "success")
-            agents_list.append(
-                {
-                    "agent_id": name,
-                    "agent_name": name,
-                    "total_executions": total,
-                    "successful_executions": success,
-                    "failed_executions": total - success,
-                    "success_rate": success / total if total else 0.0,
-                    "average_duration": sum(m.duration_ms for m in ms) / total / 1000.0
-                    if total
-                    else 0.0,
-                }
-            )
-
-        # Workflows: by workflow_id
-        wf_counts: dict[str, list] = defaultdict(list)
-        for m in recent:
-            wf_counts[m.workflow_id].append(m)
-
-        workflows_list = []
-        for wf_id, ms in wf_counts.items():
-            total = len(ms)
-            success = sum(1 for m in ms if m.status == "success")
-            workflows_list.append(
-                {
-                    "workflow_id": wf_id,
-                    "workflow_name": wf_id,
-                    "total_executions": total,
-                    "successful_executions": success,
-                    "failed_executions": total - success,
-                    "success_rate": success / total if total else 0.0,
-                    "average_duration": sum(m.duration_ms for m in ms) / total / 1000.0
-                    if total
-                    else 0.0,
-                }
-            )
-
-        # System: try cpu/mem/disk from ResourceMonitor
-        cpu_usage = memory_usage = disk_usage = 0.0
-        try:
-            from ...core.resource_monitor import ResourceMonitor
-
-            mon = ResourceMonitor()
-            res = mon.get_current_metrics()
-            cpu_usage = getattr(res, "cpu_percent", 0.0) or 0.0
-            memory_usage = getattr(res, "memory_percent", 0.0) or 0.0
-            disk_usage = getattr(res, "disk_percent", 0.0) or 0.0
-        except Exception:
-            pass
-
-        return {
-            "timestamp": now.isoformat(),
-            "system": {
-                "timestamp": now.isoformat(),
-                "total_agents": len(agents_list),
-                "active_workflows": 0,
-                "completed_workflows_today": completed_today,
-                "failed_workflows_today": failed_today,
-                "average_workflow_duration": avg_duration,
-                "cpu_usage": cpu_usage,
-                "memory_usage": memory_usage,
-                "disk_usage": disk_usage,
-            },
-            "agents": agents_list,
-            "workflows": workflows_list,
-        }
-    except Exception:
-        return None
-
-
-def handle_health_check_command(
-    check_name: str | None = None,
-    output_format: str = "text",
-    save: bool = True,
-    project_root: Path | None = None,
-) -> None:
-    """
-    Handle health check command.
-
-    Args:
-        check_name: Optional specific check to run
-        output_format: Output format (json or text)
-        save: Whether to save results to metrics storage
-        project_root: Project root directory
-    """
-    project_root = project_root or Path.cwd()
-
-    # Initialize registry and register all checks
-    registry = HealthCheckRegistry()
-    registry.register(EnvironmentHealthCheck(project_root=project_root))
-    registry.register(AutomationHealthCheck(project_root=project_root))
-    registry.register(ExecutionHealthCheck(project_root=project_root))
-    registry.register(Context7CacheHealthCheck(project_root=project_root))
-    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
-    registry.register(OutcomeHealthCheck(project_root=project_root))
-
-    # Initialize orchestrator
-    metrics_collector = HealthMetricsCollector(project_root=project_root)
-    orchestrator = HealthOrchestrator(
-        registry=registry, metrics_collector=metrics_collector, project_root=project_root
-    )
-
-    # Run checks
-    feedback = get_feedback()
-    feedback.format_type = output_format
-    operation_desc = f"Running health check: {check_name}" if check_name else "Running all health checks"
-    feedback.start_operation("Health Check", operation_desc)
-    
-    if check_name:
-        check_names = [check_name]
-        feedback.running(f"Initializing check: {check_name}...", step=1, total_steps=3)
-    else:
-        check_names = None
-        feedback.running("Discovering health checks...", step=1, total_steps=3)
-
-    feedback.running("Executing health checks...", step=2, total_steps=3)
-    results = orchestrator.run_all_checks(check_names=check_names, save_metrics=save)
-    feedback.running("Collecting results...", step=3, total_steps=3)
-    feedback.clear_progress()
-
-    # Build summary
-    summary = {}
-    if results:
-        healthy_count = sum(1 for r in results.values() if r and r.status == "healthy")
-        total_count = len([r for r in results.values() if r])
-        summary["checks_run"] = total_count
-        summary["healthy"] = healthy_count
-        summary["degraded"] = sum(1 for r in results.values() if r and r.status == "degraded")
-        summary["unhealthy"] = sum(1 for r in results.values() if r and r.status == "unhealthy")
-
-    # Format output
-    if output_format == "json":
-        output = {
-            "checks": {
-                name: {
-                    "status": result.status,
-                    "score": result.score,
-                    "message": result.message,
-                    "details": result.details,
-                    "remediation": (
-                        result.remediation
-                        if isinstance(result.remediation, list)
-                        else [result.remediation]
-                        if result.remediation
-                        else None
-                    ),
-                }
-                for name, result in results.items()
-                if result
-            }
-        }
-        # Merge summary into output
-        if summary:
-            output = {**output, "summary": summary}
-        feedback.output_result(output, message="Health checks completed")
-    else:
-        # Text output
-        feedback.success("Health checks completed")
-        warnings = []
-        for name, result in sorted(results.items()):
-            if not result:
-                continue
-
-            status_symbol = {
-                "healthy": "[OK]",
-                "degraded": "[WARN]",
-                "unhealthy": "[FAIL]",
-            }.get(result.status, "[?]")
-
-            print(f"\n[{status_symbol}] {name.upper()}: {result.status} ({result.score:.1f}/100)")
-            print(f"   {result.message}")
-            
-            if result.status != "healthy":
-                warnings.append(f"{name}: {result.message}")
-
-            if result.details:
-                # Show key metrics
-                key_metrics = []
-                for key in [
-                    "total_executions",
-                    "success_rate",
-                    "hit_rate",
-                    "total_files",
-                    "average_score",
-                ]:
-                    if key in result.details:
-                        value = result.details[key]
-                        if isinstance(value, float):
-                            if key == "success_rate" or key == "hit_rate":
-                                key_metrics.append(f"{key}: {value:.1f}%")
-                            else:
-                                key_metrics.append(f"{key}: {value:.1f}")
-                        else:
-                            key_metrics.append(f"{key}: {value}")
-
-                if key_metrics:
-                    print(f"   Metrics: {' | '.join(key_metrics)}")
-
-            if result.remediation:
-                if isinstance(result.remediation, list):
-                    if len(result.remediation) > 0:
-                        print(f"   Remediation: {result.remediation[0]}")
-                elif isinstance(result.remediation, str):
-                    print(f"   Remediation: {result.remediation}")
-        
-        if warnings:
-            for warning_msg in warnings:
-                feedback.warning(warning_msg)
-
-
-def handle_health_dashboard_command(
-    output_format: str = "text", project_root: Path | None = None
-) -> None:
-    """
-    Handle health dashboard command.
-
-    Args:
-        output_format: Output format (json or text)
-        project_root: Project root directory
-    """
-    project_root = project_root or Path.cwd()
-
-    # Initialize registry and register all checks
-    registry = HealthCheckRegistry()
-    registry.register(EnvironmentHealthCheck(project_root=project_root))
-    registry.register(AutomationHealthCheck(project_root=project_root))
-    registry.register(ExecutionHealthCheck(project_root=project_root))
-    registry.register(Context7CacheHealthCheck(project_root=project_root))
-    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
-    registry.register(OutcomeHealthCheck(project_root=project_root))
-
-    # Initialize dashboard
-    metrics_collector = HealthMetricsCollector(project_root=project_root)
-    orchestrator = HealthOrchestrator(
-        registry=registry, metrics_collector=metrics_collector, project_root=project_root
-    )
-    dashboard = HealthDashboard(orchestrator=orchestrator)
-
-    # Render dashboard
-    feedback = get_feedback()
-    feedback.format_type = output_format
-    feedback.start_operation("Health Dashboard", "Generating health dashboard visualization")
-    feedback.running("Collecting health metrics...", step=1, total_steps=3)
-    feedback.running("Generating dashboard...", step=2, total_steps=3)
-    feedback.running("Rendering dashboard output...", step=3, total_steps=3)
-    
-    if output_format == "json":
-        output = dashboard.render_json()
-        feedback.clear_progress()
-        feedback.output_result(output, message="Health dashboard generated")
-    else:
-        output = dashboard.render_text()
-        feedback.clear_progress()
-        feedback.success("Health dashboard generated")
-        print(output)
-
-
-def handle_health_metrics_command(
-    check_name: str | None = None,
-    status: str | None = None,
-    days: int = 30,
-    output_format: str = "text",
-    project_root: Path | None = None,
-) -> None:
-    """
-    Handle health metrics command.
-
-    Args:
-        check_name: Optional check name to filter
-        status: Optional status to filter
-        days: Number of days to look back
-        output_format: Output format (json or text)
-        project_root: Project root directory
-    """
-    project_root = project_root or Path.cwd()
-    collector = HealthMetricsCollector(project_root=project_root)
-
-    # Get metrics
-    feedback = get_feedback()
-    feedback.format_type = output_format
-    operation_desc = f"Collecting metrics{f' for {check_name}' if check_name else ''}"
-    feedback.start_operation("Health Metrics", operation_desc)
-    feedback.running("Querying metrics database...", step=1, total_steps=3)
-    
-    metrics = collector.get_metrics(check_name=check_name, status=status, days=days, limit=1000)
-    feedback.running("Calculating summary statistics...", step=2, total_steps=3)
-    summary = collector.get_summary(days=days)
-    feedback.running("Formatting results...", step=3, total_steps=3)
-    feedback.clear_progress()
-
-    if output_format == "json":
-        output = {
-            "summary": summary,
-            "metrics": [m.to_dict() for m in metrics],
-        }
-        feedback.output_result(output, message="Health metrics retrieved")
-    else:
-        # Text output
-        feedback.success("Health metrics retrieved")
-        print(f"\nHealth Metrics Summary (last {days} days)")
-        print("=" * 70)
-        print(f"Total checks: {summary['total_checks']}")
-        print(f"Average score: {summary['average_score']:.1f}/100")
-        print(f"\nBy status:")
-        for status_name, count in summary["by_status"].items():
-            print(f"  {status_name}: {count}")
-
-        if summary["by_check"]:
-            print(f"\nBy check:")
-            for check_name, check_data in summary["by_check"].items():
-                print(f"  {check_name}:")
-                print(f"    Count: {check_data['count']}")
-                print(f"    Average score: {check_data['average_score']:.1f}/100")
-                print(f"    Latest status: {check_data['latest_status']}")
-                print(f"    Latest score: {check_data['latest_score']:.1f}/100")
-
-        if metrics:
-            print(f"\nRecent metrics (showing up to 10):")
-            for metric in metrics[:10]:
-                print(f"  {metric.check_name}: {metric.status} ({metric.score:.1f}/100) - {metric.timestamp}")
-
-
-def handle_health_trends_command(
-    check_name: str,
-    days: int = 7,
-    output_format: str = "text",
-    project_root: Path | None = None,
-) -> None:
-    """
-    Handle health trends command.
-
-    Args:
-        check_name: Check name to analyze trends for
-        days: Number of days to analyze
-        output_format: Output format (json or text)
-        project_root: Project root directory
-    """
-    project_root = project_root or Path.cwd()
-    collector = HealthMetricsCollector(project_root=project_root)
-
-    # Get trends
-    feedback = get_feedback()
-    feedback.format_type = output_format
-    feedback.start_operation("Health Trends", f"Analyzing health trends for {check_name}")
-    feedback.running("Loading historical data...", step=1, total_steps=3)
-    
-    trends = collector.get_trends(check_name=check_name, days=days)
-    feedback.running("Calculating trends...", step=2, total_steps=3)
-    feedback.running("Generating trend report...", step=3, total_steps=3)
-    feedback.clear_progress()
-
-    if output_format == "json":
-        output = {
-            "check_name": check_name,
-            "days": days,
-            "trends": trends,
-        }
-        feedback.output_result(output, message="Health trends analyzed")
-    else:
-        # Text output
-        feedback.success("Health trends analyzed")
-        print(f"\nHealth Trends for '{check_name}' (last {days} days)")
-        print("=" * 70)
-        print(f"Direction: {trends['direction']}")
-        print(f"Score change: {trends['score_change']:+.1f} points")
-
-        if trends["status_changes"]:
-            print(f"\nStatus changes:")
-            for status, change in trends["status_changes"].items():
-                if change != 0:
-                    print(f"  {status}: {change:+d}")
-
-
-def handle_health_usage_command(args: object) -> None:
-    """
-    Handle health usage subcommand (formerly analytics).
-    Dispatches to dashboard, agents, workflows, trends, or system using AnalyticsDashboard.
-    """
-    from ...core.analytics_dashboard import AnalyticsDashboard
-
-    dashboard = AnalyticsDashboard()
-    sub = getattr(args, "usage_subcommand", "dashboard")
-    if sub == "show":
-        sub = "dashboard"
-    fmt = getattr(args, "format", "text")
-
-    if sub == "dashboard":
-        data = dashboard.get_dashboard_data()
-        if fmt == "json":
-            format_json_output(data)
-        else:
-            print("\n" + "=" * 60)
-            print("Usage / Analytics Dashboard")
-            print("=" * 60)
-            print(f"\nSystem Status (as of {data['timestamp']}):")
-            sys_data = data["system"]
-            print(f"  Total Agents: {sys_data['total_agents']}")
-            print(f"  Active Workflows: {sys_data['active_workflows']}")
-            print(f"  Completed Today: {sys_data['completed_workflows_today']}")
-            print(f"  Failed Today: {sys_data['failed_workflows_today']}")
-            print(f"  Avg Workflow Duration: {sys_data['average_workflow_duration']:.2f}s")
-            print(f"  CPU Usage: {sys_data['cpu_usage']:.1f}%")
-            print(f"  Memory Usage: {sys_data['memory_usage']:.1f}%")
-            print(f"  Disk Usage: {sys_data['disk_usage']:.1f}%")
-            print("\nAgent Performance (Top 10):")
-            for agent in sorted(data["agents"], key=lambda x: x["total_executions"], reverse=True)[:10]:
-                print(f"  {agent['agent_name']}: {agent['total_executions']} executions, "
-                      f"{agent['success_rate']*100:.1f}% success, {agent['average_duration']:.2f}s avg")
-            print("\nWorkflow Performance:")
-            for wf in sorted(data["workflows"], key=lambda x: x["total_executions"], reverse=True)[:10]:
-                print(f"  {wf['workflow_name']}: {wf['total_executions']} executions, "
-                      f"{wf['success_rate']*100:.1f}% success")
-    elif sub == "agents":
-        metrics = dashboard.get_agent_performance(agent_id=getattr(args, "agent_id", None))
-        if fmt == "json":
-            format_json_output(metrics)
-        else:
-            for agent in metrics:
-                print(f"{agent['agent_name']}: {agent['total_executions']} executions, "
-                      f"{agent['success_rate']*100:.1f}% success")
-    elif sub == "workflows":
-        metrics = dashboard.get_workflow_performance(workflow_id=getattr(args, "workflow_id", None))
-        if fmt == "json":
-            format_json_output(metrics)
-        else:
-            for wf in metrics:
-                print(f"{wf['workflow_name']}: {wf['total_executions']} executions, "
-                      f"{wf['success_rate']*100:.1f}% success")
-    elif sub == "trends":
-        metric_type = getattr(args, "metric_type", "agent_duration")
-        days = getattr(args, "days", 30)
-        trends = dashboard.get_trends(metric_type, days=days)
-        if fmt == "json":
-            format_json_output(trends)
-        else:
-            for t in trends:
-                print(f"{t['metric_name']}: {len(t['values'])} data points")
-    elif sub == "system":
-        status = dashboard.get_system_status()
-        if fmt == "json":
-            format_json_output(status)
-        else:
-            print(f"System Status (as of {status['timestamp']}):")
-            print(f"  Total Agents: {status['total_agents']}")
-            print(f"  Active Workflows: {status['active_workflows']}")
-            print(f"  Completed Today: {status['completed_workflows_today']}")
-            print(f"  Failed Today: {status['failed_workflows_today']}")
-
-
-def handle_health_overview_command(
-    output_format: str = "text",
-    project_root: Path | None = None,
-) -> None:
-    """
-    Single 1000-foot view: health checks + usage rolled up for all subsystems.
-
-    Renders one easy-to-read report: overall health, each health check one line,
-    then usage at a glance (system, top agents, top workflows).
-    """
-    from ...core.analytics_dashboard import AnalyticsDashboard
-
-    project_root = project_root or Path.cwd()
-
-    # 1. Health checks
-    registry = HealthCheckRegistry()
-    registry.register(EnvironmentHealthCheck(project_root=project_root))
-    registry.register(AutomationHealthCheck(project_root=project_root))
-    registry.register(ExecutionHealthCheck(project_root=project_root))
-    registry.register(Context7CacheHealthCheck(project_root=project_root))
-    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
-    registry.register(OutcomeHealthCheck(project_root=project_root))
-    metrics_collector = HealthMetricsCollector(project_root=project_root)
-    orchestrator = HealthOrchestrator(
-        registry=registry,
-        metrics_collector=metrics_collector,
-        project_root=project_root,
-    )
-    health_results = orchestrator.run_all_checks(save_metrics=True)
-    overall = orchestrator.get_overall_health(health_results)
-
-    # 2. Usage (best-effort; prefer analytics, fallback to execution metrics — HM-001-S1)
-    _log = logging.getLogger(__name__)
-    usage_data = None
-    try:
-        usage_dashboard = AnalyticsDashboard()
-        usage_data = usage_dashboard.get_dashboard_data()
-    except Exception:
-        pass
-    # If analytics has no agent/workflow data, derive from execution metrics
-    fallback_used = False
-    if usage_data:
-        agents = usage_data.get("agents") or []
-        workflows = usage_data.get("workflows") or []
-        total_runs = sum(a.get("total_executions", 0) for a in agents) + sum(
-            w.get("total_executions", 0) for w in workflows
-        )
-        if total_runs == 0:
-            fallback = _usage_data_from_execution_metrics(project_root)
-            if fallback:
-                fallback_used = True
-                usage_data = fallback
-    else:
-        fallback = _usage_data_from_execution_metrics(project_root)
-        if fallback:
-            fallback_used = True
-            usage_data = fallback
-    if fallback_used and usage_data:
-        n_agents = len(usage_data.get("agents") or [])
-        n_workflows = len(usage_data.get("workflows") or [])
-        _log.info(
-            "Health overview: using execution metrics fallback (%s agents, %s workflows)",
-            n_agents, n_workflows,
-        )
-
-    # 3. Build output
-    feedback = get_feedback()
-    feedback.format_type = output_format
-
-    if output_format == "json":
-        out = {
-            "overview": {
-                "overall_health": overall,
-                "health_checks": {
-                    name: {
-                        "status": r.status,
-                        "score": r.score,
-                        "message": r.message,
-                    }
-                    for name, r in health_results.items()
-                    if r
-                },
-            },
-            "usage": usage_data,
-        }
-        format_json_output(out)
-        return
-
-    # Text: 1000-foot, great-looking, easy to read
-    width = 72
-    lines = []
-    lines.append("")
-    lines.append("=" * width)
-    lines.append("  TAPPS-AGENTS  |  HEALTH + USAGE  |  1000-FOOT VIEW")
-    lines.append("=" * width)
-    lines.append("")
-
-    # Overall health
-    status_sym = {"healthy": "[OK] ", "degraded": "[WARN]", "unhealthy": "[FAIL]", "unknown": "[?]  "}
-    sym = status_sym.get(overall["status"], "[?]  ")
-    lines.append(f"  {sym}  Overall: {overall['status'].upper()}  ({overall['score']:.1f}/100)")
-    lines.append("")
-
-    # Subsystems (health checks) - one line each
-    lines.append("  SUBSYSTEMS (health)")
-    lines.append("  " + "-" * (width - 2))
-    for name, result in sorted(health_results.items()):
-        if not result:
-            continue
-        s = status_sym.get(result.status, "[?]  ")
-        label = name.replace("_", " ").upper()
-        lines.append(f"  {s}  {label}: {result.score:.1f}/100  |  {result.message[:50]}{'...' if len(result.message) > 50 else ''}")
-    lines.append("")
-
-    # Usage at a glance
-    lines.append("  USAGE (agents & workflows)")
-    lines.append("  " + "-" * (width - 2))
-    if usage_data:
-        sys_data = usage_data.get("system", {})
-        lines.append(f"  Today: completed {sys_data.get('completed_workflows_today', 0)} workflows, failed {sys_data.get('failed_workflows_today', 0)}  |  active: {sys_data.get('active_workflows', 0)}")
-        lines.append(f"  Avg workflow duration: {sys_data.get('average_workflow_duration', 0):.1f}s  |  CPU: {sys_data.get('cpu_usage', 0):.0f}%  Mem: {sys_data.get('memory_usage', 0):.0f}%  Disk: {sys_data.get('disk_usage', 0):.0f}%")
-        agents = sorted(usage_data.get("agents", []), key=lambda x: x.get("total_executions", 0), reverse=True)[:5]
-        if agents:
-            lines.append("  Top agents (30d): " + "  |  ".join(f"{a.get('agent_name', '')}: {a.get('total_executions', 0)} runs ({a.get('success_rate', 0)*100:.0f}% ok)" for a in agents))
-        workflows = sorted(usage_data.get("workflows", []), key=lambda x: x.get("total_executions", 0), reverse=True)[:5]
-        if workflows:
-            lines.append("  Top workflows (30d): " + "  |  ".join(f"{w.get('workflow_name', '')}: {w.get('total_executions', 0)} ({w.get('success_rate', 0)*100:.0f}% ok)" for w in workflows))
-    else:
-        lines.append("  (No usage data yet. Run agents/workflows to populate.)")
-    lines.append("")
-    lines.append("=" * width)
-    lines.append("")
-
-    feedback.clear_progress()
-    print("\n".join(lines))
-
+"""
+Health command handlers.
+"""
+
+from __future__ import annotations
+
+import json
+import logging
+import sys
+from collections import defaultdict
+from datetime import UTC, datetime, timedelta
+from pathlib import Path
+
+from ...health.checks.automation import AutomationHealthCheck
+from ...health.checks.environment import EnvironmentHealthCheck
+from ...health.checks.execution import ExecutionHealthCheck
+from ...health.checks.context7_cache import Context7CacheHealthCheck
+from ...health.checks.knowledge_base import KnowledgeBaseHealthCheck
+from ...health.checks.outcomes import OutcomeHealthCheck
+from ...health.collector import HealthMetricsCollector
+from ...health.dashboard import HealthDashboard
+from ...health.orchestrator import HealthOrchestrator
+from ...health.registry import HealthCheckRegistry
+from ..feedback import get_feedback, ProgressTracker
+from .common import format_json_output
+
+
+def _usage_data_from_execution_metrics(project_root: Path) -> dict | None:
+    """
+    Build usage-like data from execution metrics when analytics is empty.
+
+    Aggregates .tapps-agents/metrics/executions_*.jsonl by today (steps/workflows),
+    by skill (agents), and by workflow_id (workflows). Returns same shape as
+    AnalyticsDashboard.get_dashboard_data() for system/agents/workflows.
+    """
+    try:
+        from ...workflow.execution_metrics import ExecutionMetricsCollector
+
+        collector = ExecutionMetricsCollector(project_root=project_root)
+        metrics = collector.get_metrics(limit=5000)
+        if not metrics:
+            return None
+
+        now = datetime.now(UTC)
+        today_start = now.replace(hour=0, minute=0, second=0, microsecond=0)
+        thirty_days_ago = now - timedelta(days=30)
+
+        # Filter to last 30 days
+        def parse_ts(ts: str) -> datetime:
+            return datetime.fromisoformat(ts.replace("Z", "+00:00"))
+
+        recent = [m for m in metrics if parse_ts(m.started_at) >= thirty_days_ago]
+        if not recent:
+            return None
+
+        today_metrics = [m for m in recent if parse_ts(m.started_at) >= today_start]
+        workflow_ids_today_success = {
+            m.workflow_id for m in today_metrics if m.status == "success"
+        }
+        workflow_ids_today_failed = {
+            m.workflow_id for m in today_metrics if m.status != "success"
+        }
+        completed_today = len(workflow_ids_today_success)
+        failed_today = len(workflow_ids_today_failed)
+        avg_duration = (
+            sum(m.duration_ms for m in recent) / len(recent) / 1000.0
+            if recent
+            else 0.0
+        )
+
+        # Agents: by skill or command
+        agent_counts: dict[str, list] = defaultdict(list)
+        for m in recent:
+            key = m.skill or m.command or "unknown"
+            agent_counts[key].append(m)
+
+        agents_list = []
+        for name, ms in agent_counts.items():
+            total = len(ms)
+            success = sum(1 for m in ms if m.status == "success")
+            agents_list.append(
+                {
+                    "agent_id": name,
+                    "agent_name": name,
+                    "total_executions": total,
+                    "successful_executions": success,
+                    "failed_executions": total - success,
+                    "success_rate": success / total if total else 0.0,
+                    "average_duration": sum(m.duration_ms for m in ms) / total / 1000.0
+                    if total
+                    else 0.0,
+                }
+            )
+
+        # Workflows: by workflow_id
+        wf_counts: dict[str, list] = defaultdict(list)
+        for m in recent:
+            wf_counts[m.workflow_id].append(m)
+
+        workflows_list = []
+        for wf_id, ms in wf_counts.items():
+            total = len(ms)
+            success = sum(1 for m in ms if m.status == "success")
+            workflows_list.append(
+                {
+                    "workflow_id": wf_id,
+                    "workflow_name": wf_id,
+                    "total_executions": total,
+                    "successful_executions": success,
+                    "failed_executions": total - success,
+                    "success_rate": success / total if total else 0.0,
+                    "average_duration": sum(m.duration_ms for m in ms) / total / 1000.0
+                    if total
+                    else 0.0,
+                }
+            )
+
+        # System: try cpu/mem/disk from ResourceMonitor
+        cpu_usage = memory_usage = disk_usage = 0.0
+        try:
+            from ...core.resource_monitor import ResourceMonitor
+
+            mon = ResourceMonitor()
+            res = mon.get_current_metrics()
+            cpu_usage = getattr(res, "cpu_percent", 0.0) or 0.0
+            memory_usage = getattr(res, "memory_percent", 0.0) or 0.0
+            disk_usage = getattr(res, "disk_percent", 0.0) or 0.0
+        except Exception:
+            pass
+
+        return {
+            "timestamp": now.isoformat(),
+            "system": {
+                "timestamp": now.isoformat(),
+                "total_agents": len(agents_list),
+                "active_workflows": 0,
+                "completed_workflows_today": completed_today,
+                "failed_workflows_today": failed_today,
+                "average_workflow_duration": avg_duration,
+                "cpu_usage": cpu_usage,
+                "memory_usage": memory_usage,
+                "disk_usage": disk_usage,
+            },
+            "agents": agents_list,
+            "workflows": workflows_list,
+        }
+    except Exception:
+        return None
+
+
+def handle_health_check_command(
+    check_name: str | None = None,
+    output_format: str = "text",
+    save: bool = True,
+    project_root: Path | None = None,
+) -> None:
+    """
+    Handle health check command.
+
+    Args:
+        check_name: Optional specific check to run
+        output_format: Output format (json or text)
+        save: Whether to save results to metrics storage
+        project_root: Project root directory
+    """
+    project_root = project_root or Path.cwd()
+
+    # Initialize registry and register all checks
+    registry = HealthCheckRegistry()
+    registry.register(EnvironmentHealthCheck(project_root=project_root))
+    registry.register(AutomationHealthCheck(project_root=project_root))
+    registry.register(ExecutionHealthCheck(project_root=project_root))
+    registry.register(Context7CacheHealthCheck(project_root=project_root))
+    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
+    registry.register(OutcomeHealthCheck(project_root=project_root))
+
+    # Initialize orchestrator
+    metrics_collector = HealthMetricsCollector(project_root=project_root)
+    orchestrator = HealthOrchestrator(
+        registry=registry, metrics_collector=metrics_collector, project_root=project_root
+    )
+
+    # Run checks
+    feedback = get_feedback()
+    feedback.format_type = output_format
+    operation_desc = f"Running health check: {check_name}" if check_name else "Running all health checks"
+    feedback.start_operation("Health Check", operation_desc)
+    
+    if check_name:
+        check_names = [check_name]
+        feedback.running(f"Initializing check: {check_name}...", step=1, total_steps=3)
+    else:
+        check_names = None
+        feedback.running("Discovering health checks...", step=1, total_steps=3)
+
+    feedback.running("Executing health checks...", step=2, total_steps=3)
+    results = orchestrator.run_all_checks(check_names=check_names, save_metrics=save)
+    feedback.running("Collecting results...", step=3, total_steps=3)
+    feedback.clear_progress()
+
+    # Build summary
+    summary = {}
+    if results:
+        healthy_count = sum(1 for r in results.values() if r and r.status == "healthy")
+        total_count = len([r for r in results.values() if r])
+        summary["checks_run"] = total_count
+        summary["healthy"] = healthy_count
+        summary["degraded"] = sum(1 for r in results.values() if r and r.status == "degraded")
+        summary["unhealthy"] = sum(1 for r in results.values() if r and r.status == "unhealthy")
+
+    # Format output
+    if output_format == "json":
+        output = {
+            "checks": {
+                name: {
+                    "status": result.status,
+                    "score": result.score,
+                    "message": result.message,
+                    "details": result.details,
+                    "remediation": (
+                        result.remediation
+                        if isinstance(result.remediation, list)
+                        else [result.remediation]
+                        if result.remediation
+                        else None
+                    ),
+                }
+                for name, result in results.items()
+                if result
+            }
+        }
+        # Merge summary into output
+        if summary:
+            output = {**output, "summary": summary}
+        feedback.output_result(output, message="Health checks completed")
+    else:
+        # Text output
+        feedback.success("Health checks completed")
+        warnings = []
+        for name, result in sorted(results.items()):
+            if not result:
+                continue
+
+            status_symbol = {
+                "healthy": "[OK]",
+                "degraded": "[WARN]",
+                "unhealthy": "[FAIL]",
+            }.get(result.status, "[?]")
+
+            print(f"\n[{status_symbol}] {name.upper()}: {result.status} ({result.score:.1f}/100)")
+            print(f"   {result.message}")
+            
+            if result.status != "healthy":
+                warnings.append(f"{name}: {result.message}")
+
+            if result.details:
+                # Show key metrics
+                key_metrics = []
+                for key in [
+                    "total_executions",
+                    "success_rate",
+                    "hit_rate",
+                    "total_files",
+                    "average_score",
+                ]:
+                    if key in result.details:
+                        value = result.details[key]
+                        if isinstance(value, float):
+                            if key == "success_rate" or key == "hit_rate":
+                                key_metrics.append(f"{key}: {value:.1f}%")
+                            else:
+                                key_metrics.append(f"{key}: {value:.1f}")
+                        else:
+                            key_metrics.append(f"{key}: {value}")
+
+                if key_metrics:
+                    print(f"   Metrics: {' | '.join(key_metrics)}")
+
+            if result.remediation:
+                if isinstance(result.remediation, list):
+                    if len(result.remediation) > 0:
+                        print(f"   Remediation: {result.remediation[0]}")
+                elif isinstance(result.remediation, str):
+                    print(f"   Remediation: {result.remediation}")
+        
+        if warnings:
+            for warning_msg in warnings:
+                feedback.warning(warning_msg)
+
+
+def handle_health_dashboard_command(
+    output_format: str = "text", project_root: Path | None = None
+) -> None:
+    """
+    Handle health dashboard command.
+
+    Args:
+        output_format: Output format (json or text)
+        project_root: Project root directory
+    """
+    project_root = project_root or Path.cwd()
+
+    # Initialize registry and register all checks
+    registry = HealthCheckRegistry()
+    registry.register(EnvironmentHealthCheck(project_root=project_root))
+    registry.register(AutomationHealthCheck(project_root=project_root))
+    registry.register(ExecutionHealthCheck(project_root=project_root))
+    registry.register(Context7CacheHealthCheck(project_root=project_root))
+    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
+    registry.register(OutcomeHealthCheck(project_root=project_root))
+
+    # Initialize dashboard
+    metrics_collector = HealthMetricsCollector(project_root=project_root)
+    orchestrator = HealthOrchestrator(
+        registry=registry, metrics_collector=metrics_collector, project_root=project_root
+    )
+    dashboard = HealthDashboard(orchestrator=orchestrator)
+
+    # Render dashboard
+    feedback = get_feedback()
+    feedback.format_type = output_format
+    feedback.start_operation("Health Dashboard", "Generating health dashboard visualization")
+    feedback.running("Collecting health metrics...", step=1, total_steps=3)
+    feedback.running("Generating dashboard...", step=2, total_steps=3)
+    feedback.running("Rendering dashboard output...", step=3, total_steps=3)
+    
+    if output_format == "json":
+        output = dashboard.render_json()
+        feedback.clear_progress()
+        feedback.output_result(output, message="Health dashboard generated")
+    else:
+        output = dashboard.render_text()
+        feedback.clear_progress()
+        feedback.success("Health dashboard generated")
+        print(output)
+
+
+def handle_health_metrics_command(
+    check_name: str | None = None,
+    status: str | None = None,
+    days: int = 30,
+    output_format: str = "text",
+    project_root: Path | None = None,
+) -> None:
+    """
+    Handle health metrics command.
+
+    Args:
+        check_name: Optional check name to filter
+        status: Optional status to filter
+        days: Number of days to look back
+        output_format: Output format (json or text)
+        project_root: Project root directory
+    """
+    project_root = project_root or Path.cwd()
+    collector = HealthMetricsCollector(project_root=project_root)
+
+    # Get metrics
+    feedback = get_feedback()
+    feedback.format_type = output_format
+    operation_desc = f"Collecting metrics{f' for {check_name}' if check_name else ''}"
+    feedback.start_operation("Health Metrics", operation_desc)
+    feedback.running("Querying metrics database...", step=1, total_steps=3)
+    
+    metrics = collector.get_metrics(check_name=check_name, status=status, days=days, limit=1000)
+    feedback.running("Calculating summary statistics...", step=2, total_steps=3)
+    summary = collector.get_summary(days=days)
+    feedback.running("Formatting results...", step=3, total_steps=3)
+    feedback.clear_progress()
+
+    if output_format == "json":
+        output = {
+            "summary": summary,
+            "metrics": [m.to_dict() for m in metrics],
+        }
+        feedback.output_result(output, message="Health metrics retrieved")
+    else:
+        # Text output
+        feedback.success("Health metrics retrieved")
+        print(f"\nHealth Metrics Summary (last {days} days)")
+        print("=" * 70)
+        print(f"Total checks: {summary['total_checks']}")
+        print(f"Average score: {summary['average_score']:.1f}/100")
+        print(f"\nBy status:")
+        for status_name, count in summary["by_status"].items():
+            print(f"  {status_name}: {count}")
+
+        if summary["by_check"]:
+            print(f"\nBy check:")
+            for check_name, check_data in summary["by_check"].items():
+                print(f"  {check_name}:")
+                print(f"    Count: {check_data['count']}")
+                print(f"    Average score: {check_data['average_score']:.1f}/100")
+                print(f"    Latest status: {check_data['latest_status']}")
+                print(f"    Latest score: {check_data['latest_score']:.1f}/100")
+
+        if metrics:
+            print(f"\nRecent metrics (showing up to 10):")
+            for metric in metrics[:10]:
+                print(f"  {metric.check_name}: {metric.status} ({metric.score:.1f}/100) - {metric.timestamp}")
+
+
+def handle_health_trends_command(
+    check_name: str,
+    days: int = 7,
+    output_format: str = "text",
+    project_root: Path | None = None,
+) -> None:
+    """
+    Handle health trends command.
+
+    Args:
+        check_name: Check name to analyze trends for
+        days: Number of days to analyze
+        output_format: Output format (json or text)
+        project_root: Project root directory
+    """
+    project_root = project_root or Path.cwd()
+    collector = HealthMetricsCollector(project_root=project_root)
+
+    # Get trends
+    feedback = get_feedback()
+    feedback.format_type = output_format
+    feedback.start_operation("Health Trends", f"Analyzing health trends for {check_name}")
+    feedback.running("Loading historical data...", step=1, total_steps=3)
+    
+    trends = collector.get_trends(check_name=check_name, days=days)
+    feedback.running("Calculating trends...", step=2, total_steps=3)
+    feedback.running("Generating trend report...", step=3, total_steps=3)
+    feedback.clear_progress()
+
+    if output_format == "json":
+        output = {
+            "check_name": check_name,
+            "days": days,
+            "trends": trends,
+        }
+        feedback.output_result(output, message="Health trends analyzed")
+    else:
+        # Text output
+        feedback.success("Health trends analyzed")
+        print(f"\nHealth Trends for '{check_name}' (last {days} days)")
+        print("=" * 70)
+        print(f"Direction: {trends['direction']}")
+        print(f"Score change: {trends['score_change']:+.1f} points")
+
+        if trends["status_changes"]:
+            print(f"\nStatus changes:")
+            for status, change in trends["status_changes"].items():
+                if change != 0:
+                    print(f"  {status}: {change:+d}")
+
+
+def handle_health_usage_command(args: object) -> None:
+    """
+    Handle health usage subcommand (formerly analytics).
+    Dispatches to dashboard, agents, workflows, trends, or system using AnalyticsDashboard.
+    """
+    from ...core.analytics_dashboard import AnalyticsDashboard
+
+    dashboard = AnalyticsDashboard()
+    sub = getattr(args, "usage_subcommand", "dashboard")
+    if sub == "show":
+        sub = "dashboard"
+    fmt = getattr(args, "format", "text")
+
+    if sub == "dashboard":
+        data = dashboard.get_dashboard_data()
+        if fmt == "json":
+            format_json_output(data)
+        else:
+            print("\n" + "=" * 60)
+            print("Usage / Analytics Dashboard")
+            print("=" * 60)
+            print(f"\nSystem Status (as of {data['timestamp']}):")
+            sys_data = data["system"]
+            print(f"  Total Agents: {sys_data['total_agents']}")
+            print(f"  Active Workflows: {sys_data['active_workflows']}")
+            print(f"  Completed Today: {sys_data['completed_workflows_today']}")
+            print(f"  Failed Today: {sys_data['failed_workflows_today']}")
+            print(f"  Avg Workflow Duration: {sys_data['average_workflow_duration']:.2f}s")
+            print(f"  CPU Usage: {sys_data['cpu_usage']:.1f}%")
+            print(f"  Memory Usage: {sys_data['memory_usage']:.1f}%")
+            print(f"  Disk Usage: {sys_data['disk_usage']:.1f}%")
+            print("\nAgent Performance (Top 10):")
+            for agent in sorted(data["agents"], key=lambda x: x["total_executions"], reverse=True)[:10]:
+                print(f"  {agent['agent_name']}: {agent['total_executions']} executions, "
+                      f"{agent['success_rate']*100:.1f}% success, {agent['average_duration']:.2f}s avg")
+            print("\nWorkflow Performance:")
+            for wf in sorted(data["workflows"], key=lambda x: x["total_executions"], reverse=True)[:10]:
+                print(f"  {wf['workflow_name']}: {wf['total_executions']} executions, "
+                      f"{wf['success_rate']*100:.1f}% success")
+    elif sub == "agents":
+        metrics = dashboard.get_agent_performance(agent_id=getattr(args, "agent_id", None))
+        if fmt == "json":
+            format_json_output(metrics)
+        else:
+            for agent in metrics:
+                print(f"{agent['agent_name']}: {agent['total_executions']} executions, "
+                      f"{agent['success_rate']*100:.1f}% success")
+    elif sub == "workflows":
+        metrics = dashboard.get_workflow_performance(workflow_id=getattr(args, "workflow_id", None))
+        if fmt == "json":
+            format_json_output(metrics)
+        else:
+            for wf in metrics:
+                print(f"{wf['workflow_name']}: {wf['total_executions']} executions, "
+                      f"{wf['success_rate']*100:.1f}% success")
+    elif sub == "trends":
+        metric_type = getattr(args, "metric_type", "agent_duration")
+        days = getattr(args, "days", 30)
+        trends = dashboard.get_trends(metric_type, days=days)
+        if fmt == "json":
+            format_json_output(trends)
+        else:
+            for t in trends:
+                print(f"{t['metric_name']}: {len(t['values'])} data points")
+    elif sub == "system":
+        status = dashboard.get_system_status()
+        if fmt == "json":
+            format_json_output(status)
+        else:
+            print(f"System Status (as of {status['timestamp']}):")
+            print(f"  Total Agents: {status['total_agents']}")
+            print(f"  Active Workflows: {status['active_workflows']}")
+            print(f"  Completed Today: {status['completed_workflows_today']}")
+            print(f"  Failed Today: {status['failed_workflows_today']}")
+
+
+def handle_health_overview_command(
+    output_format: str = "text",
+    project_root: Path | None = None,
+) -> None:
+    """
+    Single 1000-foot view: health checks + usage rolled up for all subsystems.
+
+    Renders one easy-to-read report: overall health, each health check one line,
+    then usage at a glance (system, top agents, top workflows).
+    """
+    from ...core.analytics_dashboard import AnalyticsDashboard
+
+    project_root = project_root or Path.cwd()
+
+    # 1. Health checks
+    registry = HealthCheckRegistry()
+    registry.register(EnvironmentHealthCheck(project_root=project_root))
+    registry.register(AutomationHealthCheck(project_root=project_root))
+    registry.register(ExecutionHealthCheck(project_root=project_root))
+    registry.register(Context7CacheHealthCheck(project_root=project_root))
+    registry.register(KnowledgeBaseHealthCheck(project_root=project_root))
+    registry.register(OutcomeHealthCheck(project_root=project_root))
+    metrics_collector = HealthMetricsCollector(project_root=project_root)
+    orchestrator = HealthOrchestrator(
+        registry=registry,
+        metrics_collector=metrics_collector,
+        project_root=project_root,
+    )
+    health_results = orchestrator.run_all_checks(save_metrics=True)
+    overall = orchestrator.get_overall_health(health_results)
+
+    # 2. Usage (best-effort; prefer analytics, fallback to execution metrics — HM-001-S1)
+    _log = logging.getLogger(__name__)
+    usage_data = None
+    try:
+        usage_dashboard = AnalyticsDashboard()
+        usage_data = usage_dashboard.get_dashboard_data()
+    except Exception:
+        pass
+    # If analytics has no agent/workflow data, derive from execution metrics
+    fallback_used = False
+    if usage_data:
+        agents = usage_data.get("agents") or []
+        workflows = usage_data.get("workflows") or []
+        total_runs = sum(a.get("total_executions", 0) for a in agents) + sum(
+            w.get("total_executions", 0) for w in workflows
+        )
+        if total_runs == 0:
+            fallback = _usage_data_from_execution_metrics(project_root)
+            if fallback:
+                fallback_used = True
+                usage_data = fallback
+    else:
+        fallback = _usage_data_from_execution_metrics(project_root)
+        if fallback:
+            fallback_used = True
+            usage_data = fallback
+    if fallback_used and usage_data:
+        n_agents = len(usage_data.get("agents") or [])
+        n_workflows = len(usage_data.get("workflows") or [])
+        _log.info(
+            "Health overview: using execution metrics fallback (%s agents, %s workflows)",
+            n_agents, n_workflows,
+        )
+
+    # 3. Build output
+    feedback = get_feedback()
+    feedback.format_type = output_format
+
+    if output_format == "json":
+        out = {
+            "overview": {
+                "overall_health": overall,
+                "health_checks": {
+                    name: {
+                        "status": r.status,
+                        "score": r.score,
+                        "message": r.message,
+                    }
+                    for name, r in health_results.items()
+                    if r
+                },
+            },
+            "usage": usage_data,
+        }
+        format_json_output(out)
+        return
+
+    # Text: 1000-foot, great-looking, easy to read
+    width = 72
+    lines = []
+    lines.append("")
+    lines.append("=" * width)
+    lines.append("  TAPPS-AGENTS  |  HEALTH + USAGE  |  1000-FOOT VIEW")
+    lines.append("=" * width)
+    lines.append("")
+
+    # Overall health
+    status_sym = {"healthy": "[OK] ", "degraded": "[WARN]", "unhealthy": "[FAIL]", "unknown": "[?]  "}
+    sym = status_sym.get(overall["status"], "[?]  ")
+    lines.append(f"  {sym}  Overall: {overall['status'].upper()}  ({overall['score']:.1f}/100)")
+    lines.append("")
+
+    # Subsystems (health checks) - one line each
+    lines.append("  SUBSYSTEMS (health)")
+    lines.append("  " + "-" * (width - 2))
+    for name, result in sorted(health_results.items()):
+        if not result:
+            continue
+        s = status_sym.get(result.status, "[?]  ")
+        label = name.replace("_", " ").upper()
+        lines.append(f"  {s}  {label}: {result.score:.1f}/100  |  {result.message[:50]}{'...' if len(result.message) > 50 else ''}")
+    lines.append("")
+
+    # Usage at a glance
+    lines.append("  USAGE (agents & workflows)")
+    lines.append("  " + "-" * (width - 2))
+    if usage_data:
+        sys_data = usage_data.get("system", {})
+        lines.append(f"  Today: completed {sys_data.get('completed_workflows_today', 0)} workflows, failed {sys_data.get('failed_workflows_today', 0)}  |  active: {sys_data.get('active_workflows', 0)}")
+        lines.append(f"  Avg workflow duration: {sys_data.get('average_workflow_duration', 0):.1f}s  |  CPU: {sys_data.get('cpu_usage', 0):.0f}%  Mem: {sys_data.get('memory_usage', 0):.0f}%  Disk: {sys_data.get('disk_usage', 0):.0f}%")
+        agents = sorted(usage_data.get("agents", []), key=lambda x: x.get("total_executions", 0), reverse=True)[:5]
+        if agents:
+            lines.append("  Top agents (30d): " + "  |  ".join(f"{a.get('agent_name', '')}: {a.get('total_executions', 0)} runs ({a.get('success_rate', 0)*100:.0f}% ok)" for a in agents))
+        workflows = sorted(usage_data.get("workflows", []), key=lambda x: x.get("total_executions", 0), reverse=True)[:5]
+        if workflows:
+            lines.append("  Top workflows (30d): " + "  |  ".join(f"{w.get('workflow_name', '')}: {w.get('total_executions', 0)} ({w.get('success_rate', 0)*100:.0f}% ok)" for w in workflows))
+    else:
+        lines.append("  (No usage data yet. Run agents/workflows to populate.)")
+    lines.append("")
+    lines.append("=" * width)
+    lines.append("")
+
+    feedback.clear_progress()
+    print("\n".join(lines))
+