meta-health

Aggregate telemetry and workflow data to generate Markdown health dashboards.

2|3|Updated Oct 20, 2025

Search Tags:#metrics #meta-learning #telemetry #ROI #health-dashboard #agent-evolution

Ready to run

npx skills add https://github.com/psd401/psd-claude-coding-system --skill meta-health

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: meta-health
Source: https://github.com/psd401/psd-claude-coding-system/tree/main/plugins/psd-claude-coding-system/skills/meta-health
Command: npx skills add https://github.com/psd401/psd-claude-coding-system --skill meta-health

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Teams building AI-powered software need continuous visibility into system health, ROI, and learning progress. This Skill provides an automated dashboard that aggregates telemetry, meta-learning data, and agent metrics to reveal the health of the compound engineering system.

Core Features & Use Cases

Unified health dashboard: collects telemetry.json, compound_history.json, experiments.json, agent_variants.json, and workflow_graph.json to compute health metrics.
Velocity and ROI insights: tracks developer velocity, time saved, and compound ROI to guide prioritization.
Agent evolution and patterns: analyzes agent variants and workflow patterns to identify improvements and optimization opportunities.
Publication and summaries: supports publishing dashboards and email summaries with optional flags.

Quick Start

Use the meta-health skill to generate a current health dashboard. Example: /meta-health --output meta/health-YYYYMMDD.md --publish

Frequently Asked Questions about meta-health

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

FAQPage Schema

How do I automate system health dashboards for AI-powered software projects?▼

System health dashboards aggregate telemetry, experiments, and agent metrics to track velocity and ROI. Meta-health generates Markdown dashboards by loading telemetry.json, compound_history.json, experiments.json, agent_variants.json, and workflow_graph.json, then computes and renders health metrics on weekly or monthly intervals.

What metrics does a health dashboard track for compound engineering systems?▼

Health dashboards track developer velocity, time saved, compound ROI, agent evolution patterns, and workflow optimization opportunities. Meta-health reveals these through aggregated telemetry and meta-learning data to guide prioritization and identify system improvements.

Can I publish and email health dashboard summaries automatically?▼

Yes. Meta-health supports optional publish and email summary flags when generating dashboards. Use --publish to enable dashboard distribution alongside Markdown output to meta/health-YYYYMMDD.md.

What input data files do I need before generating a health dashboard?▼

Meta-health requires five JSON files: meta/telemetry.json for system metrics, meta/compound_history.json for learning history, meta/experiments.json for trial results, meta/agent_variants.json for agent versions, and meta/workflow_graph.json for execution patterns.

How often should I regenerate health dashboards to track ROI?▼

Meta-health supports monthly or weekly intervals depending on your project cadence. Regular regeneration keeps velocity, time savings, and risk metrics current for informed prioritization decisions.