ChatGPT can now analyze your GitHub repositories
OpenAI has introduced a GitHub connector for its ChatGPT deep research feature, allowing developers to ask questions about codebases and engineering documents. The connector is now available in beta and enables ChatGPT to link to GitHub, enhancing its ability to compile thorough research reports.
The new feature is available for ChatGPT Plus, Pro, and Team users and will be rolled out over the next few days. Enterprise and Edu support are expected to follow soon, according to an OpenAI spokesperson. With this capability, users can break down product specs into technical tasks and dependencies, summarize code structure and patterns, and understand how to implement new APIs using real code examples.
ChatGPT deep research respects an organization’s GitHub settings, ensuring users only see content they are allowed to view and codebases that have been explicitly shared with ChatGPT. OpenAI emphasizes that the feature is a potential time saver, not a replacement for experts, as there is a risk that ChatGPT deep research may hallucinate or provide inaccurate information.
“I often hear that users find ChatGPT’s deep research agent so valuable that they want it to connect to their internal sources, in addition to the web,” OpenAI Head of Business Products Nate Gonzalez wrote in a blog post on LinkedIn. “Today we’re introducing our first connector.”
This development is part of OpenAI’s broader efforts to enhance its AI-powered tools for assistive coding. The company has recently unveiled an open-source coding tool for terminals called Codex CLI and upgraded the ChatGPT desktop app to read code in various developer-focused coding apps. OpenAI reportedly reached an agreement to buy AI-powered coding assistant Windsurf for $3 billion.
In related news, OpenAI has launched fine-tuning options for developers to customize its newer models for specific applications. Developers can now fine-tune OpenAI’s o4-mini “reasoning” model using reinforcement fine-tuning, which involves task-specific grading to improve the model’s performance. Fine-tuning is also available for the GPT-4.1 nano model, with verified organizations able to fine-tune o4-mini and all paying developers able to fine-tune GPT-4.1 nano.
OpenAI began requiring verification for certain models and developer features in April to prevent abuse, a process that involves submitting an ID and other identity documents.