skill-health

Analyze JSONL skill invocation logs to report performance metrics and trends.

127|22|Updated Feb 20, 2026
One-click install
npx skills add https://github.com/flonat/flonat-research --skill skill-health-flonat
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-health
Source: https://github.com/flonat/flonat-research/tree/main/skills/skill-health
Command: npx skills add https://github.com/flonat/flonat-research --skill skill-health-flonat

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill provides a comprehensive dashboard for monitoring the health of skills, including invocation counts, success rates, failure patterns, and overall status. It reads JSONL logs and qualitative observations to report on skill performance without modifying them.

Core Features & Use Cases

  • Skill Invocation Metrics: Track the number of times a skill is invoked and its success rate.
  • Health Status Analysis: Assess the health of skills by analyzing failure patterns and trends over time.
  • Qualitative Observations: Review qualitative observations for any issues or concerns.
  • Trend Detection: Identify trends such as declining or improving health status over time.
  • Label Distribution: Analyze the distribution of labels associated with skills to identify potential issues.
  • Failures-Only Mode: Filter and display only error/partial entries in the period, grouped by skill.
  • Label Reconciliation: Reconcile logged labels against the on-disk skill list to ensure accurate health reporting.
  • Dead Skill Detection: Identify skills with zero invocations over a long period as dormant for further review.
  • Weekly Review Trigger: Suggest running the weekly review protocol if it has been more than 7 days since the last review.

Quick Start

Run the skill health assessment and present results for all skills.

Frequently Asked Questions about skill-health

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

FAQPage Schema
How do I monitor skill invocation metrics and success rates from JSONL logs?

To monitor skill invocation metrics, this Skill reads JSONL logs to calculate invocation counts, success rates, and failure patterns. It processes the log data using Python scripts to generate a detailed health dashboard.

What is the best way to detect declining health trends and failure patterns in skills?

Detecting declining health trends involves analyzing invocation logs and qualitative observations to identify failure patterns over time. This assessment tool provides trend detection to pinpoint deteriorating skill performance.

How do I find dormant skills with zero invocations over a long period?

Finding dormant skills with zero invocations is handled by the dead skill detection feature, which identifies skills lacking log activity over a long period. This flags them for further review and potential removal.

Can I filter invocation logs to display only error and partial failure entries?

Yes, you can filter invocation logs to display only error and partial entries using the failures-only mode. This groups error instances by skill to isolate and review specific performance issues quickly.

Do I need Python to analyze label distribution and reconcile logged skill labels?

Yes, you need Python to analyze label distribution and reconcile logged labels against the on-disk skill list. The scripts process the JSONL data to ensure accurate health reporting and identify potential label issues.

When should I run a weekly review protocol for skill health assessment?

You should run a weekly review protocol for skill health assessment if it has been more than 7 days since the last review. The monitoring dashboard suggests triggering this protocol to maintain regular performance checks.