Aggregate Aeon skill run snapshots into weekly fleet analytics reports.
npx skills add https://github.com/aaronjmars/aeon --skill skill-analytics-aaronjmars
Please help me install this Agent Skill. Skill: skill-analytics Source: https://github.com/aaronjmars/aeon/tree/main/skills/skill-analytics Command: npx skills add https://github.com/aaronjmars/aeon --skill skill-analytics-aaronjmars
It converts raw per-run skill outcomes into a single weekly, fleet-level report so operators can quickly see what’s running most, what’s failing most, what’s quietly skipping, and which scheduled skills never fired.
Run the skill with the default 7-day window to generate a fleet analytics report covering your latest skill runs and anomalies.
High-intent search queries and answers about installing and using this skill.
Fleet monitoring detects silent failures by aggregating run snapshots and cron scheduling expectations, identifying cron-enabled skills with zero runs in the selected time window. This flags scheduled skills that appear active but never actually fired.
Exit taxonomy distribution mines daily logs to categorize skill outcomes into buckets like ok, quiet, skip, partial, and error. This distinguishes degraded quiet correctness from true failures, revealing how frequently skills end in skip-class outcomes versus hard errors.
Run the analytics skill with the default 7-day window to aggregate run snapshots, cron streak signals, and exit taxonomy data. It produces a ranked top-runner view, failure-rate breakdown, and anomaly-gated verdict written to articles and dashboard JSON outputs.
Yes, fleet-wide ranking produces a view of every skill with runs in the window, including success and failure counts plus last status. Anomaly-first reporting applies significance gating to flag degraded success rates and prioritizes the strongest verdict for operator attention.
Aggregate cron-state streak signals alongside run snapshots to detect scheduled skills with zero runs in the window. This best-effort approach compares expected cron schedules against actual execution frequency to surface missed or silently failing runs.
Silent scheduled detection identifies cron-enabled skills that appear active in configuration but have zero runs in the window. This occurs when skills are misconfigured, blocked, or failing to trigger, requiring aggregation of cron scheduling expectations against actual run snapshots.
anajuliabitGenerate fleet-level weekly analytics reports ranking Aeon skills by run volume and exit taxonomy.
swarm-ai-researchSynthesize weekly agent health, community growth, and economic signals into an operator scorecard.
swarm-ai-researchAggregate and rank Aeon skills by weekly run counts for fleet analytics.
Atrium-HermesAggregate Aeon skill run data into fleet-wide rankings and anomaly reports.
