A project template that implements the essential directories for scaling AI engineering work:
- AI Docs: A persistent knowledge repository for AI tools
- Specs: Detailed plans and specifications for features
- .claude: Reusable prompts for agentic coding tools
- Examples: Practical workflows demonstrating how to use the template with AI tools
ai-code-template/
├── ai-docs/
│ ├── third-party/
│ ├── conventions.md
│ └── architecture.md
├── specs/
│ └── template.md
├── .claude/
│ └── commands/
│ ├── context-prime.prompt
│ └── feature-prime.prompt
├── examples/
│ ├── README.md
│ └── feature_development.md
├── tools/
│ └── pocket-pick/
│ ├── db.js
│ ├── cli.js
│ └── README.md
└── README.md
ai_docs/: Store third-party API documentation, integration details, custom patterns, and conventionsspecs/: Store detailed plans and specifications for features.claude/commands/: Store reusable prompts for agentic coding toolsexamples/: Practical workflow examples showing how to use the template with AI tools
- Clone this repository as a starting point for your new project
- Add project-specific documentation to the AI docs directory
- Write comprehensive spec plans in the specs directory
- Set up reusable prompts in the .claude/commands directory
- Prime the AI coding tool with the context of the codebase using the context priming prompt
- Write agentic plans for desired features in the specs directory
- Have the AI coding tool implement the plans
- Review the examples directory for practical workflows demonstrating these steps
- AI tools can understand your codebase better and faster
- Scale your engineering work across codebases and time
- Write self-validating loops in prompts to expand into full codebases
- Create detailed plans for your AI tools to implement
- Reuse prompts across sessions with your AI coding tools
For more information, see the documentation in each directory.