One-click install
npx skills add https://github.com/swarm-ai-research/aeon --skill skill-analytics-swarm-ai-research
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-analytics
Source: https://github.com/swarm-ai-research/aeon/tree/main/skills/skill-analytics
Command: npx skills add https://github.com/swarm-ai-research/aeon --skill skill-analytics-swarm-ai-research

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Fleet-wide analytics of Aeon skill runs, ranking skills by a specified window and surfacing deploys of success, failures, and anomalies to improve observability.

Core Features & Use Cases

  • Fleet-wide ranking of skill runs by window (e.g., 7d/168h) to identify busiest and most unreliable skills.
  • Anomaly detection by exit taxonomy distribution and silent-scheduled runs to flag degraded fleet health.
  • Dashboard-ready outputs including a fleet article, JSON spec, and a dashboard payload for observability.

Quick Start

Run the skill-analytics to generate the latest fleet analytics report for the default 7-day window.

Frequently Asked Questions about skill-analytics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I monitor cron-driven skill health across an entire fleet?

You monitor fleet health by aggregating weekly run counts and analyzing exit taxonomy from logs. This process cross-references cron schedules and memory state to surface top runners, failure rates, and silent-scheduled skills within a configurable window.

What is exit taxonomy analysis and how does it detect fleet anomalies?

Exit taxonomy analysis evaluates log distributions from skill runs to detect degraded fleet health. By categorizing exit statuses, it flags silent-scheduled runs and identifies non-firing skills, providing anomaly detection for fleet-wide observability.

How do I generate a dashboard JSON spec from run history data?

Generate a dashboard JSON spec by reading per-skill run snapshots and cross-referencing cron schedules. The output includes a dashboard payload and a fleet article that rank skills by a configurable window to improve observability.

Can I identify silent-scheduled and non-firing skills within a custom time window?

Yes, you can identify silent-scheduled and non-firing skills within a configurable window like the default 7-day or 168-hour period. The analytics tool cross-references cron schedules with memory state to flag these anomalies.

Does fleet analytics work without external dependencies for observability?

Fleet analytics operates without external dependencies, reading per-skill run snapshots internally. It analyzes exit taxonomy and memory state directly to output a dashboard payload and fleet article for self-contained observability.