Common Questions
What is Git Digest?
Git Digest converts a Git repository into a simple text representation that can be easily fed into Large Language Models (LLMs). It allows you to extract the code content from any public GitHub repository and format it in a way that's optimized for LLM consumption.
Why would I use Git Digest?
When working with AI tools like ChatGPT, Claude, or other LLMs, you often need to provide context about your codebase. Git Digest helps you convert your repository into a format that's easily digestible by these AI systems, allowing them to better understand your code structure and provide more relevant assistance.
How does Git Digest work?
Git Digest fetches the repository content via GitHub's API, filters files based on your criteria, processes the code to ensure it stays within token limits, and produces a text representation. All processing happens in your browser, ensuring your code remains private.
Is there a size limit for repositories?
Yes. The default size limit is set to 50KB to ensure reasonable processing times and to stay within token limits for most LLMs. You can adjust this setting if needed, but be aware that very large repositories may cause performance issues.
Can I use this with private repositories?
Currently, Git Digest only works with public GitHub repositories. Support for private repositories or other Git providers may be added in the future.
What file types are included by default?
By default, Git Digest focuses on code files and excludes documentation, build artifacts, and binary files. The default exclusion patterns filter out common non-code files like images, videos, and large data files. You can customize these filters as needed.