This page intentionally omits informal “research dumps” and generic anti-patterns lists—favor the linked handbooks and standards below for actionable guidance.
Core handbooks
- Beyond Vibe Coding A practical long-form guide to AI-assisted development as a spectrum, from fast prototyping to production-ready engineering.
- AI Code Guide A free GitHub handbook that consolidates AI coding workflows, prompting practices, planning, testing, and safety guidelines.
- Agentic Coding A reference-manual style course for engineers covering agentic workflows, planning, testing, debugging, and code review with AI.
Official documentation stacks
- OpenAI developer docs OpenAI’s documentation ecosystem covering best practices, prompting, workflows, skills, AGENTS.md usage, and coding agents.
- Claude Code tutorials Anthropic’s practical guides for using Claude as a coding partner, including debugging, refactoring, and codebase analysis.
Community long-form guide
- The Ultimate Vibe Coding Guide (Reddit) A large community-written guide describing real-world workflows for using AI IDEs effectively on larger codebases.
Courses and training
- Generative AI for Software Development DeepLearning.AI course on using LLMs as pair programmers, testers, and design assistants while maintaining engineering discipline.
Principles and standards
- Agentic Coding Principles & Practices A framework defining six principles and 28 practices for building reliable systems with coding agents.
- AGENTS.md An open convention for a machine-readable project guide: context, commands, and rules for coding agents.
Tools
Opinionated workflows and skill libraries you can install or fork—useful when you want a structured “team in a box” on top of your editor or agent.- gstack Garry Tan’s open-source Claude Code workflow: slash-command roles, browser automation, and a long-lived stack for high-throughput AI-assisted shipping.
- Superpowers An agentic skills framework and software development methodology—skills, patterns, and practices you can wire into compatible agents and IDEs.