context-engineering

Create persistent project rules files for AI coding tools.

Updated Jun 1, 2026
One-click install
npx skills add https://github.com/j-cogburn/d3 --skill context-engineering-j-cogburn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: context-engineering
Source: https://github.com/j-cogburn/d3/tree/main/tests/project-t01/.d3/skills/context-engineering
Command: npx skills add https://github.com/j-cogburn/d3 --skill context-engineering-j-cogburn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Poorly configured context leads to low-quality AI agent output: insufficient context causes hallucinations and ignored project conventions, while excessive irrelevant context makes agents lose focus and drift from task requirements. This Skill provides a structured framework to curate optimal context for AI-assisted development workflows.

Core Features & Use Cases

  • Context Hierarchy Setup: Organize context from persistent project-wide rules to transient task-specific information to ensure agents always access relevant data.
  • Cross-Tool Rules Files: Create compatible rules files for Claude Code (CLAUDE.md), Cursor (.cursorrules), GitHub Copilot, and other AI coding tools to enforce project standards.
  • Ambiguity Resolution: Use proven patterns to surface and resolve context conflicts and missing requirements instead of letting agents make incorrect assumptions.
  • Use Case: When starting a new software project, use this Skill to set up a rules file that encodes your tech stack, code conventions, and project boundaries, eliminating repeated corrections for agent mistakes.

Quick Start

Use the context-engineering skill to create a tailored rules file for your AI coding tool that includes your project's tech stack, code conventions, and boundaries to improve agent output quality.

Frequently Asked Questions about context-engineering

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

FAQPage Schema
Why does my AI coding agent hallucinate or ignore project conventions during software development?

AI coding agents hallucinate or ignore conventions due to poorly curated context, where insufficient project rules or excessive irrelevant information causes focus loss. Context engineering resolves this by structuring optimal context hierarchies.

How do I create rules files for Claude Code, Cursor, and GitHub Copilot to enforce coding standards?

You create cross-tool rules files like CLAUDE.md and .cursorrules by organizing a context hierarchy from persistent project-wide rules to transient task-specific data, ensuring AI coding tools enforce project standards consistently.

What is context engineering for AI agents and when do I need it?

Context engineering for AI agents is the structured curation of context configurations to eliminate hallucinations and focus loss. You need it during AI-assisted coding workflows like session initialization, task switching, and project onboarding.

How do I resolve context conflicts and missing requirements when using AI coding tools?

Resolve context conflicts and missing requirements in AI coding tools by applying proven ambiguity resolution patterns that surface hidden issues, preventing agents from making incorrect assumptions about your software development project.

Can I use context engineering to manage long conversation history and task switching in AI-assisted coding?

Yes, context engineering manages long conversation history and task switching by organizing context from persistent project-wide rules to transient task-specific information, ensuring AI agents maintain focus and access relevant data across sessions.

What is the best way to set up context for a new AI-assisted software project?

The best way to set up context for a new AI-assisted software project is to encode your tech stack, code conventions, and project boundaries into a tailored rules file, establishing a clear context hierarchy to guide agent behavior.