A curated collection of high-quality instruction templates and prompt guides to help developers write better code prompts — especially for AI-powered coding assistants like Gemini CLI, GitHub Copilot, and more.
Welcome to Code Instructions Kit — your go-to repository for well-crafted instruction templates aimed at improving developer workflows with AI coding tools. Whether you're using Gemini CLI or GitHub Copilot, writing clear and effective prompts can drastically enhance your productivity and code quality.
This repo provides framework-specific, best-practice instruction files that help you define coding style, component design patterns, testing strategies, and repository conventions — so your AI coding assistant understands exactly what you expect.
The rise of AI-assisted coding is reshaping software development, but the quality of output depends heavily on the quality of instructions and context provided. Unfortunately, crafting these prompts and instructions from scratch can be tedious and inconsistent across teams.
This project aims to:
-
Provide ready-to-use, standardized instruction templates for popular frontend frameworks and tooling combinations (React, Next.js, Tailwind, Vue, Vite, etc.)
-
Share best practices distilled from industry experience for component design, testing, accessibility, and code style
-
Enable developers at all levels to accelerate their AI-assisted coding workflows with minimal setup
-
Foster a community-driven collection where contributors can share their own templates, improvements, and ideas
This kit is designed to seamlessly integrate into your projects, offering tangible benefits:
- Rapid AI Assistant Setup: Quickly configure your AI coding assistant with battle-tested instruction files, ensuring it understands your project's specific conventions from day one.
- Enforce Best Practices: Leverage pre-defined templates to guide your AI in generating code that adheres to industry-standard component design, testing, and accessibility principles.
- Tailored Code Generation: Customize prompts and guidelines to precisely match your team's unique coding style, architectural patterns, and project-specific requirements, leading to more accurate and idiomatic AI-generated code.
- Standardized Collaboration: Establish a shared understanding of coding standards across your team, fostering consistency and reducing friction in open-source or enterprise projects.
Follow these steps to empower your AI coding assistant with project-specific guidelines:
-
Clone the Repository:
git clone https://github.com/abhinavkumar985/code-instructions-kit.git cd code-instructions-kit
-
Identify Your Stack: Browse the folders (e.g.,
react/
,node/
,stacks/react-next-tailwind/
) to find theGEMINI.md
and.gemini-guidelines/
directory that best matches your project's technology stack. -
Integrate into Your Project:
- For New Projects: Copy the entire
GEMINI.md
file and its corresponding.gemini-guidelines/
directory directly into your new project's root folder. - For Existing Projects:
- Copy the
GEMINI.md
file and the.gemini-guidelines/
directory into your project's root. - If you already have a
GEMINI.md
or similar configuration, merge the relevant sections from this kit'sGEMINI.md
into your existing one. Ensure the.gemini-guidelines/
directory is placed alongside yourGEMINI.md
file.
- Copy the
Example Structure in Your Project Root:
- For New Projects: Copy the entire
your-project/
├── GEMINI.md
├── .gemini-guidelines/
│ ├── overview.md
│ └── component-design.md
└── src/
└── components
-
Customize and Extend:
- Open the copied
GEMINI.md
and the files within.gemini-guidelines/
. - Modify their content to reflect your project's unique coding standards, architectural decisions, and team preferences. This is crucial for tailoring the AI's output.
- Add new guideline files to
.gemini-guidelines/
as needed, and remember to reference them in yourGEMINI.md
.
- Open the copied
-
Leverage with Your AI Assistant:
- For Gemini CLI: The CLI automatically detects
GEMINI.md
and its referenced guidelines in your project root. Simply start interacting with the CLI, and it will use these instructions to inform its responses and code generation. - For Other AI Assistants (e.g., GitHub Copilot): Copy the relevant
content from the
.gemini-guidelines/
files into your assistant's configuration or prompt context.
- For Gemini CLI: The CLI automatically detects
To get the most out of this kit and your AI coding assistant:
- Adapt, Don't Just Adopt: Treat these templates as robust starting points. Actively customize them to precisely fit your project's unique context, team culture, and technical requirements.
- Maintain as a Living Document: Your codebase evolves, and so should your
guidelines. Regularly review and update your
GEMINI.md
and.gemini-guidelines/
files to reflect changes in your project's architecture, dependencies, or best practices. - Contribute Back: Your improvements and new templates are invaluable! Share them with the community via Pull Requests to help this resource grow and benefit others.
- Stay Updated: Follow this repository to receive notifications on new additions, updated best practices, and enhancements to the kit.
We welcome and encourage contributions from the community! Your insights and additions are vital to making this kit a comprehensive and invaluable resource for developers.
If you have:
- Instruction templates for other frameworks, libraries, or tooling combinations.
- Improvements or refinements for existing files.
- Suggestions for new sections, topics, or best practices.
Please open an issue or submit a pull request. Let's collaborate to build the ultimate resource for AI-assisted coding!
This project is open source under the MIT License.
If you find this repo helpful, please ⭐ star it and share it with your developer network. Follow me for more insights on frontend development, AI tools, and open source contributions.
🐙 GitHub: https://github.com/abhinavkumar985
🐦 Twitter: https://x.com/abhinav985
💼 LinkedIn: https://www.linkedin.com/in/abhinavkumar985/
🌐 Portfolio: https://abhinavkumar985.github.io/portfolio/about
✍️ Medium: https://medium.com/@abhinavkumar985
Thank you for visiting Code Instructions Kit — happy coding! 🎉