addy-context-engineering

Configures rules files and curates context to improve AI agent output quality.

Updated Aug 21, 2026
One-click install
npx skills add https://github.com/TylerSimons1127/vibe --skill addy-context-engineering-tylersimons1127
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: addy-context-engineering
Source: https://github.com/TylerSimons1127/vibe/tree/main/skills/addy-context-engineering
Command: npx skills add https://github.com/TylerSimons1127/vibe --skill addy-context-engineering-tylersimons1127

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI coding agents hallucinate APIs, ignore project conventions, and degrade in quality when given too little or too much context. This Skill provides a structured methodology for curating exactly what the agent sees at each stage of a session. ## Core Features & Use Cases - Context Hierarchy Framework: Organizes context into five levels from persistent rules files (CLAUDE.md, AGENTS.md, .cursorrules) down to transient conversation history. - Context Packing Strategies: Provides Brain Dump, Selective Include, and Hierarchical Summary patterns for loading task-relevant files, specs, and examples. - Confusion Management: Defines explicit patterns for surfacing conflicting specs, missing requirements, and inline planning instead of silently guessing. - Use Case: When starting a new feature, you create a CLAUDE.md rules file with your tech stack, commands, and conventions, then load only the relevant spec section and source files so the agent follows existing patterns instead of inventing new ones. ## Quick Start Set up a CLAUDE.md rules file for my project covering the tech stack, build commands, code conventions, and boundaries.

Frequently Asked Questions about addy-context-engineering

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

FAQPage Schema
How do I set up a CLAUDE.md rules file for my project?▼

Create a CLAUDE.md file at your project root covering tech stack, build and test commands, code conventions, and boundaries like never committing secrets. Include one short example of a well-written component in your style so the agent follows your patterns.

How much context should I give an AI coding agent per task?▼

Aim for under 2,000 lines of focused context per task. Loading more than 5,000 lines of non-task-specific context causes the agent to lose focus, so include only relevant source files, the applicable spec section, and one pattern example.

What is the equivalent of CLAUDE.md for Cursor or Copilot?▼

Cursor uses .cursorrules or .cursor/rules/*.md files, Windsurf uses .windsurfrules, GitHub Copilot uses .github/copilot-instructions.md, and OpenAI Codex uses AGENTS.md. All serve the same purpose of persistent project-wide rules.

Why does my AI agent ignore project conventions?▼

Agents ignore conventions when no rules file exists or when conventions are only implicit knowledge. If a rule is not written down, the agent cannot follow it, so document your stack, commands, and patterns in a persistent rules file.

When should I start a fresh agent session instead of continuing?▼

Start a fresh session when switching between major features or when conversation history accumulates stale context. Long sessions cause the agent to reference outdated patterns or deleted code, degrading output quality over time.