Coder is an AI-powered code generation tool designed to help developers ship code faster. It leverages artificial intelligence to generate pull requests based on project issues, streamlining the development process and boosting productivity.
- AI-Generated Pull Requests: Automatically create PRs based on issue descriptions and context.
- GitHub Integration: Seamless connection with your GitHub repositories to manage issues more effectively.
- Template Management: Create and utilize templates for consistent code generation and task automation.
- Instruction System: Define custom instructions to guide AI-generated code outputs tailored to specific needs.
- Machine Learning Model Adaptation: Utilize machine learning models to modify existing code to improve performance and readability.
- Multi-Project Support: Manage multiple projects within a singular advanced workspace.
- Simple and Advanced Modes: Choose between straightforward setups or a feature-rich development environment, enabling customization according to user skill levels.
- Real-time Code Improvement: Use AI to analyze and suggest optimizations for existing code directly from repositories.


Check out the latest demo of Coder in action: Coder Demo
Coder offers two setup modes: Simple and Advanced. Choose the one that best fits your needs.
Follow these instructions to set up a simple environment that supports AI features.
-
Clone the Repository
git clone https://github.com/bastosmichael/coder.git cd coder
-
Install Dependencies
npm install
-
Set Up Environment Variables Copy the
.env.example
to.env.development.local
and fill in the respective values. Pull any existing values from Vercel so your local environment matches production:cp .env.example .env.development.local vercel env pull .env.development.local
Required variables:
NEXT_PUBLIC_APP_MODE=simple
ANTHROPIC_API_KEY=
OPENAI_API_KEY=
DATABASE_URL=
GITHUB_PAT=
-
Run the Application
npm run dev
For a full-fledged development environment, follow the setup instructions for Advanced Mode.
- Additional setup steps specific to enabling more complex AI features will be provided in future updates.
-
Create a Postgres database in Neon or open the Neon console from your Vercel project.
-
Make sure the connection string is stored in your Vercel
DATABASE_URL
environment variable. -
Pull the latest environment variables locally and run the migrations with Drizzle:
vercel env pull .env.development.local npm run db:migrate
This ensures your local database schema matches the one deployed on Vercel.
To maximize productivity with Coder, follow these steps to utilize its AI features effectively:
- Create a Workspace: Start by creating a new workspace to organize your projects effectively.
- Add a Project: Within your workspace, create a new project and connect it to a GitHub repository.
- Generate AI Suggestions: After creating an issue in the project, utilize AI to generate code suggestions and improvements. This can be done easily through the application interface.
- Create an Issue: Describe the feature or bug fix you want to implement, and let the AI generate a corresponding pull request.
- Review and Merge PR: Use the AI suggestions, make necessary adjustments, and merge the pull request into your main branch.
To deploy Coder to Vercel and utilize its AI capabilities:
-
Fork the Coder repository to your GitHub account.
-
Sign up for a Vercel account if you haven't already.
-
Click the button below to start the deployment process:
-
Follow the prompts to configure your deployment, ensuring all required environment variables are set.
-
After deployment, set up your database and run migrations:
npx vercel env pull .env.development.local npm run db:migrate
For more detailed deployment instructions, including advanced configurations, please refer to our deployment guide.
We welcome contributions to Coder! Here's how you can help:
- Fork the repository.
- Create a new branch for your feature or bug fix.
- Make your changes and commit them with a clear, descriptive message.
- Push your changes to your fork.
- Submit a pull request to the main Coder repository.
Please read our Contribution Guidelines for more details on our code of conduct, branch naming conventions, and pull request process.
Q: How does Coder generate code? A: Coder uses advanced AI models to analyze your project structure, issue descriptions, and custom instructions to generate contextually appropriate code.
Q: Is my code safe and private? A: Yes, Coder takes security seriously. We do not store your code, and all processing is done securely. However, please review our privacy policy for more details.
Q: Can I use Coder with private repositories? A: Yes, Coder supports both public and private GitHub repositories.
For more frequently asked questions, visit our FAQ page.
Coder is open-source software licensed under the MIT license.
Coder is built using various open-source libraries and tools. A special thanks to all contributors and maintainers, including:
- Next.js
- React
- Tailwind CSS
- Drizzle ORM
- Buildware
- All contributors who helped enhance the AI capabilities in this project.