Can this skill be automated for team environments?

Yes, it supports a --json flag and specific exit codes (0 for Healthy, 1 for Needs Attention, 2 for Critical), allowing for easy integration into CI/CD pipelines.

How does this differ from the maintenance skill?

The health check skill is an auditing tool that identifies problems, while the maintenance skill is used to actually update nodes and fix the issues found during the audit.

What should I do if my health status is CRITICAL?

A CRITICAL status indicates validation failures or more than 50% stale nodes. You should address the specific failures listed or run the maintenance skill before starting new work.

How long does the quick health check take to run?

The quick check is optimized for speed and typically completes in under 30 seconds, making it perfect for non-blocking session starts.

What defines a healthy Intent Layer?

A healthy Intent Layer contains validated nodes with no structural errors, documentation refreshed within the last 30 days, and at least 80% directory coverage.

Intent Layer Health Check

Name: Intent Layer Health Check
Author: orban

byorban

•

分析与监控

Audits and validates the integrity of your Intent Layer documentation to ensure high-quality AI context and project awareness.

The Intent Layer Health Check skill provides a robust auditing system for the Intent Layer framework within Claude Code. It automatically analyzes your project's AGENTS.md files to identify stale documentation, missing coverage in key directories, and structural inconsistencies that might degrade AI performance. By performing quick or deep validations, it helps developers maintain a healthy context layer, ensuring Claude has the most accurate and up-to-date architectural understanding before starting complex coding tasks.

主要功能

01Automated staleness detection for documentation older than 90 days

02Deep consistency analysis of sections across sibling documentation nodes

033 GitHub stars

04Status-based reporting with actionable recommendations and CI-friendly exit codes

05Coverage gap identification for directories missing mandatory AGENTS.md files

06Rapid validation of root and child nodes in under 30 seconds

使用场景

01Identifying outdated documentation during large-scale refactors or codebase migrations

02Validating project context at the start of a new Claude Code session to catch issues early

03Integrating into CI/CD pipelines to enforce documentation standards and architectural integrity

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add orban/intent-layer intent-layer-health

For use in Claude.ai and ChatGPT

Download Skill