domain-context

Read DOMAIN.md at session start and propose tiered domain updates.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/factory-bucket/customware-ai --skill domain-context
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: domain-context
Source: https://github.com/factory-bucket/customware-ai/tree/main/skills/domain-context
Command: npx skills add https://github.com/factory-bucket/customware-ai --skill domain-context

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Progressive Domain Crystallization enables AI assistants to build and maintain a living, evolving domain knowledge base for any business application, ensuring terminology, entities, and rules are consistently applied across sessions.

Core Features & Use Cases

  • Read and apply a structured DOMAIN.md at session start and use progressive disclosure to load Tier 2/3 as needed.
  • Flag unknown terms and unclear relationships inline, proposing human-reviewable updates.
  • Maintain a sandboxed update process where changes are proposed, not auto-applied, preserving knowledge integrity.
  • Supports single-file and split-domain modes, routing updates to the correct detail files for escalation.

Quick Start

Load an existing DOMAIN.md for the project at the repository root, or initialize a new domain context using the provided template, then begin your session.

Frequently Asked Questions about domain-context

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

FAQPage Schema
How do I maintain a consistent domain knowledge base for AI across multiple sessions?

You can maintain a domain knowledge base across AI sessions by loading a structured DOMAIN.md file at session start. This ensures terminology and rules are consistently applied by reading the project context automatically.

What is progressive disclosure for domain context and how does it work?

Progressive disclosure for domain context is a tiered architecture that loads domain knowledge incrementally. It starts with a base DOMAIN.md and expands into Tier 2 or Tier 3 detail files only as needed by the AI session.

How do I propose updates to a domain knowledge base without auto-applying changes?

You can propose updates to a domain knowledge base using a sandboxed workflow that flags unknown terms inline. Changes are routed to a dedicated review section rather than modifying content directly, preserving knowledge integrity.

Can I use a single-file DOMAIN.md for smaller projects and split it later?

Yes, you can use a single-file DOMAIN.md for smaller projects and split it later. The system supports both single-file and split-domain modes, routing updates to the correct detail files for future escalation.

Do I need a valid SKILL.md frontmatter to manage domain context?

Yes, you need a valid SKILL.md frontmatter with a name and description to manage domain context. This frontmatter is required alongside a tiered DOMAIN.md structure to define the knowledge base parameters.