ironbots-fleet-status-monitor

Use when monitoring the Ironbots fleet execution state, checking scheduled task telemetry across background workers, detecting bot failures, or generating aggregated health status reports.

SKILL.md

GitHub

ironbots-fleet-status-monitor

Aggregates execution status and live telemetry across all 9 Ironbots fleet tasks, detects runtime errors and PGID threshold violations, and renders unified fleet health summaries.

When to Use

  • Daily scheduled fleet health reporting and telemetry aggregation
  • On-demand fleet state checks across autonomous bot processes
  • Triage and investigation of failing or degraded Ironbots scheduled tasks

Core Fleet Tasks

The monitor tracks the following 9 canonical fleet tasks:

  1. Notebook-Ingester
  2. KB-Sentinel
  3. TRM-Bot
  4. Watchlist-Miner
  5. Daemon-Healer
  6. CI-Watchdog
  7. TRM-Drive-Sync
  8. Storage-Pruner
  9. Ironbots-Reporter

Trigger & Usage

# Query live fleet health status
invoke ironbots-fleet-status-monitor { "action": "status" }

# Run full telemetry aggregation and alert detection
invoke ironbots-fleet-status-monitor { "action": "aggregate", "verbose": true }

Input Schema

interface SkillInput {
  action: "status" | "aggregate" | "diagnose";
  statusFeedDir?: string;       // Default: _status-feed/
  taskFilter?: string[];        // Optional subset of bot tasks to query
  verbose?: boolean;            // Enable detailed output
}

Output Schema

interface TaskTelemetry {
  taskId: string;
  status: "HEALTHY" | "DEGRADED" | "FAILED" | "UNKNOWN";
  lastRun?: string;
  exitCode?: number;
  message?: string;
  details?: Record<string, unknown>;
}

interface SkillOutput {
  status: "success" | "warning" | "error";
  fleetScore: number;           // 0 to 100
  totalTasks: number;
  healthyCount: number;
  failedCount: number;
  tasks: TaskTelemetry[];
  alerts: string[];
  timestamp: string;
}

Error Handling

Code Message Handler Escalation
STATUS_FEED_MISSING Telemetry directory not accessible Fallback to direct process query Inspect _status-feed/ mount
TASK_NOT_FOUND Specified task ID not recognized Skip or warn Verify task name in REQUIRED_FLEET_TASKS
TELEMETRY_CORRUPT Telemetry report JSON unparseable Mark task as UNKNOWN Inspect log format

Quick Reference

Action Purpose Typical Latency
status Lightweight query of cached task reports < 50ms
aggregate Deep scan of all 9 bot telemetry reports + scoring < 200ms
diagnose Detailed error trace of failing fleet bots < 500ms