Agent Patterns for AI Agent Development
A reference site for experienced developers leveling up with AI coding assistants. Covers single concepts per page with tool-agnostic principles and practical examples from GitHub Copilot and Claude Code.
Article intelligence
Key points
- Tool-agnostic patterns for AI agent development
- Each page focuses on one concept: what, how, when, watch out
- Includes examples from GitHub Copilot and Claude Code
- Organized into sections like Foundations, Patterns, Engineering, etc.
Why it matters
This matters because tool-agnostic patterns for AI agent development.
Technical impact
May affect model selection, inference cost, product capability, and evaluation benchmarks.
Patterns
Memory & State
Control & Orchestration
Reliability
Harness & Tools
Code-Health-Gated LLM Tier Routing
Externalization in LLM Agents
Issue Requirements Preprocessing: Structured Input Before Code Generation
Engineering
Reference
Bootstrap
Audit
Content Techniques
Technical
Measurement & Strategy
Tags
Agent Patterns¶
Patterns and techniques for experienced developers leveling up with AI coding assistants.
This is a reference site, not a tutorial. Every page covers a single concept — what it is, how it works, when to use it, and what to watch out for. Tool-agnostic principles with practical examples from GitHub Copilot and Claude Code.
Sections¶
Foundations — Context Engineering and Instructions
Patterns — Agent Design, Multi-Agent, Anti-Patterns
Engineering — Tool Engineering, Code Review, Verification, Security, Observability
GEO — Generative Engine Optimization for technical docs
Workflows — End-to-end workflows for agent-assisted development
Reference — Standards, Human Factors, Emerging, Fallacies
Tools — Claude Code, GitHub Copilot
Concept Map — All content grouped by theme
Tags — Browse all content by topic