skill-analytics

Generate fleet-level weekly analytics reports ranking Aeon skills by run volume and exit taxonomy.

6|2|Updated May 21, 2026
One-click install
npx skills add https://github.com/anajuliabit/aeon --skill skill-analytics-anajuliabit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-analytics
Source: https://github.com/anajuliabit/aeon/tree/main/skills/skill-analytics
Command: npx skills add https://github.com/anajuliabit/aeon --skill skill-analytics-anajuliabit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and assets (resource) components.

What problem does it solve?

Skill analytics eliminates the difficulty of understanding fleet-wide skill reliability by producing a single ranked report of what ran, what failed, what silently skipped, and how each skill’s exit taxonomy behaved over a time window.

Core Features & Use Cases

  • Fleet-level ranking: Summarizes every Aeon skill’s run volume and pass/fail outcome side-by-side for a configurable lookback window.
  • Silent-schedule detection: Flags enabled cron-driven skills that did not run at all during the window while ignoring on-demand workflows (like workflow_dispatch or reactive triggers).
  • Exit taxonomy insight: Detects “quiet but correct” behavior (e.g., SKIP_UNCHANGED, NEW_INFO, SKIP_QUIET) so operators can distinguish health from silent degradation.
  • Significance-gated notifications: Triggers operator alerts only when anomalies exist, reducing noise while keeping coverage for issues like repeated failures and low success rates.

Quick Start

Run the skill-analytics report for the last 7 days by setting the window to 168 hours (or another integer like 72) so it generates the latest fleet view and only notifies you if anomalies are detected.

Frequently Asked Questions about skill-analytics

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

FAQPage Schema
How do I monitor scheduled GitHub Actions for silent skips and fleet-wide reliability?

Fleet monitoring for scheduled GitHub Actions requires correlating run snapshots with cron schedules to detect silent skips. This skill produces a ranked analytics report summarizing run volume, success rates, and silent degradation across all enabled cron-driven workflows over a configurable lookback window.

What is exit taxonomy anomaly detection in autonomous agent observability?

Exit taxonomy anomaly detection identifies "quiet but correct" outcomes like SKIP_UNCHANGED or SKIP_QUIET in autonomous agent observability. This mechanism flags silent degradation by scanning daily logs for exit markers, allowing operators to distinguish healthy behavior from hidden failures in scheduled-skill workflows.

How do I generate a weekly fleet monitoring report for Aeon skills?

Generate a weekly fleet monitoring report by setting the analytics window to 168 hours. The skill fetches skill-runs JSON snapshots, correlates them with cron scheduling data, and ranks every Aeon skill by run volume, success rate, and anomaly flags for the specified lookback period.

Can I get anomaly alerts for low success rates without receiving notifications for healthy skills?

Yes, significance-gated notifications trigger operator alerts only when anomalies are detected. By applying significance-gated logic, the system reduces noise while maintaining coverage for repeated failures, low success rates, and silent schedule skips, ignoring healthy skills entirely.

Does this skill flag on-demand workflows that did not run during the lookback window?

No, silent-schedule detection specifically targets enabled cron-driven skills while ignoring on-demand workflows like workflow_dispatch or reactive triggers. This ensures fleet monitoring focuses exclusively on scheduled-skill observability anomalies rather than on-demand execution gaps.

Why does my Aeon skill show a successful run status but no actual processing output?

Successful runs with no output often reflect "quiet success" outcomes introduced by updated exit taxonomies. The analytics report detects exit markers like SKIP_UNCHANGED and NEW_INFO, helping operators interpret these silent states and distinguish correct behavior from silent degradation.