schema-drift-playbook

Classify drifting fields and draft a coordinated schema relaxation plan.

Updated Nov 18, 2025
One-click install
npx skills add https://github.com/cajias/claude-skills --skill schema-drift-playbook
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: schema-drift-playbook
Source: https://github.com/cajias/claude-skills/tree/main/plugins/dev-process/skills/schema-drift-playbook
Command: npx skills add https://github.com/cajias/claude-skills --skill schema-drift-playbook

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you resolve schema drift when validators are stricter than the data your tools and agents actually produce. It prevents the cycle where fixing only the data, only the spec, or only the producer leaves the problem returning later.

Core Features & Use Cases

  • Field Classification: Separates required fields into strict, recommended, and either-or categories so you can decide what truly must stay required.
  • Spec Relaxation: Guides you to update validators with dual-acceptance logic so both canonical and simplified shapes pass safely.
  • Migration Workflow: Recommends an idempotent CLI migration with dry-run defaults, JSON output, provenance preservation, and audit-friendly sentinels.
  • Upstream Producer Fixes: Shows how to update skills, templates, or generation paths so drift does not reappear after migration.
  • Use Case: A knowledge base lint job fails across dozens of existing records because common fields are missing; this Skill helps coordinate the validator change, backfill, and producer update together.

Quick Start

Ask the assistant to analyze the validator, classify the drifting fields, and draft a coordinated relaxation, migration, and producer-fix plan.

Frequently Asked Questions about schema-drift-playbook

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

FAQPage Schema
Why does schema drift keep returning after I fix validation failures in my data?

Schema drift returns when only the data, spec, or producer is fixed in isolation. Resolving it permanently requires coordinated spec relaxation, idempotent data migration, and upstream producer fixes to prevent the mismatch from reappearing.

How do I relax strict validators without breaking existing generated records?

To relax strict validators safely, classify fields into strict, recommended, and either-or categories, then apply dual-acceptance logic so both canonical and simplified data shapes pass validation without breaking existing records.

What is the best way to backfill missing required fields during a schema migration?

The best way to backfill missing fields is an idempotent CLI migration with dry-run defaults, JSON output, provenance preservation, and audit-friendly sentinels to safely coordinate the validator change and data update.

Can I fix frontmatter validation failures across multiple scripts and agents at once?

Yes, you can fix frontmatter validation failures across scripts and agents by coordinating spec relaxation with upstream producer fixes, ensuring templates and generation paths output the required fields to stop recurring drift.

How do I update upstream producers to prevent schema drift in generated records?

Update upstream producers by modifying skills, templates, and generation paths to output the canonical data shape. This ensures generated records match the relaxed validator spec and prevents schema drift from returning.