addy-context-engineering

What problem does it solve?

Optimizes how an AI agent sees and uses information by deliberately curating persistent and task-specific context to prevent hallucinations, drift, and misalignment.

Core Features & Use Cases

• Hierarchical context loading: organize rules, architecture docs, and relevant source files from persistent project-wide context to task-scoped context.
• Rules-driven guidance: load AGENTS.md or equivalent rules containers to enforce conventions and boundaries across sessions.
• Context layering for sessions and tasks: support persistent rules plus per-feature specs and file references to adapt to changing work scopes.
• Use Case: Starting a new development session with project conventions applied automatically to ensure consistent behavior.

Quick Start

Create a project-wide AGENTS.md with your conventions and load it as persistent context before starting any task.