health

Audit AI agent configurations and project maintainability for drift and broken hooks.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

AI-assisted coding environments silently accumulate configuration drift, broken hooks, missing verifiers, and context rot across sessions. This Skill provides a systematic, budget-aware health audit that identifies misaligned layers in agent setups and AI maintainability risks before they cause wasted tokens or broken automation.

Core Features & Use Cases

  • Agent Config Health: Detects instruction drift across Claude, Codex, and Pi configurations, validates hooks and MCP server setups, checks permission deny-floors, and audits skill supply-chain provenance.
  • AI Maintainability Health: Audits project context surfaces, verifier coverage, hotspot ownership, broken documentation references, and concentrated TODO/FIXME markers that predict future AI coding drift.
  • Tier-Aware Analysis: Adapts audit depth based on project complexity (Simple/Standard/Complex) and automatically escalates to deep audits with parallel inspector subagents when critical ambiguities or explicit deep-audit triggers are detected.

Quick Start

Use the health skill to run a summary audit of the current project's agent configuration and AI coding maintainability.

Frequently Asked Questions about health

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

FAQPage Schema
How do I audit AI agent config for instruction drift and broken hooks?

AI agent configuration audits detect instruction drift, broken hooks, and missing verifiers across Claude, Codex, and Pi environments. The audit collects agent surfaces, parses configs, and produces prioritized health reports with copy-pasteable remediation actions.

What is AI maintainability context rot and how does it affect coding environments?

AI maintainability context rot is the silent accumulation of broken documentation references, concentrated TODO markers, and misaligned layers in project setups. It causes wasted tokens and broken automation by degrading AI coding accuracy across sessions.

Can I check MCP server setups and verifier coverage for complex projects?

Yes, you can check MCP server setups and verifier coverage for complex projects. Tier-aware analysis adapts audit depth based on project complexity and automatically escalates to deep audits using parallel inspector subagents when critical ambiguities are detected.

Do I need python3 and bash to run a project health audit?

Yes, you need bash, python3, and standard Unix tooling to run a project health audit. These dependencies are required to collect agent surfaces, parse configurations, and generate prioritized health reports.

How do I fix missing verifiers and permission deny-floors in Claude configurations?

To fix missing verifiers and permission deny-floors in Claude configurations, run a health audit to identify the misaligned layers. The audit generates copy-pasteable remediation actions that systematically restore verifier coverage and validate permission settings.