Awesome Reviewers turns code-review feedback into reusable, AI-ready prompts 

How do you capture thousands of hours of code review feedback, and turn it into reusable, AI‑ready prompts? That’s what the folks at AI code review startup Baz have done with Awesome Reviewers, a public library of prompts distilled from real-world pull request comments across some of the top open source repositories. 

Code review feedback is like tribal knowledge – hard-earned lessons that rarely get documented, but shape how teams write and maintain quality code. It’s these lessons that Awesome Reviewers capture, converting them into prompts that teach the AI agents that are increasingly embedded in software development workflows.

“AI helps us write code faster than ever, but reviewing that code remains a bottleneck,” Baz co-founder and CEO Guy Eisenkot said. “For many engineering teams and open source projects, review is still inconsistent, manual, and repetitive. Standards are enforced every day, often through scattered review comments, institutional knowledge, or buried documentation.” 

Awesome Reviewers draws from more than a thousand open source projects, such as Next.js, LangChain, and FastAPI, identifying common review patterns such as style fixes, security checks, and performance tips. 

One such prompt relates to configuration usage in Node.js web framework Fastify, instructing AI code review systems to ensure configuration options are explicitly declared and properly documented, with clear examples to prevent errors and improve code clarity. Developers can then hit the Copy Prompt button and paste it into any AI code review tool that lets them customize the instructions, be that Cursor, Claude Code, or Codex. Or, if they are also Baz users, they can hit the Deploy to Baz button instead, which automatically adds the prompt into their workspace for Baz’s AI to use in pull request reviews. 

AI Code Review Prompts - Awesome Reviews by Baz

And so, similar to projects like DeepWiki and Context7, Awesome Reviewers captures unstructured developer knowledge and turns it into a reusable, structured, AI-readable format. 

It’s worth noting that Awesome Reviewers is an open source project itself, available on GitHub under an Apache 2.0 license. Through this, developers can access the raw prompt data and create deeper integrations. For example, a team could build an automated workflow that pulls relevant prompts from the GitHub repo, feeds them into their code review system or LLM, runs checks on every new pull request, and posts inline suggestions based on the results.

Prompts for Python, TypeScript, and Go

At the time of writing Awesome Reviewers features more than 470 prompts spanning 15 programming languages, though the core focus initially centers on Python, TypeScript, and Go. Eisenkot acknowledged that the library isn’t quite perfect, as some of the prompts may be too narrow or too broad. “But they’re real, and they work,” he said.

Responses from the community so far have generally been positive, with one person underscoring the value in getting real, human-like feedback early in the development process. Another was also curious as to whether the prompts transfer equally well across different programming languages.

Elsewhere, one engineer noted that the experience could be improved with smarter ways to connect prompts directly to coding tools such as Cursor, reducing the need to manually search for appropriate feedback.

That said, because Awesome Reviewers is open source, developers have the freedom to build these kinds of custom integrations themselves with a bit of elbow grease. And if truth be told, Baz might be more inclined to reserve the most seamless features – like 1-click prompt deployment – for its own platform.