yuzhichang/kunlun

Kunlun is a code search engine designed to find any piece of code across massive amounts of source files almost instantly. If your company has hundreds of code repositories with millions of lines of code spread across different teams, finding a specific function or variable can be like finding a needle in a haystack. This tool makes that search fast and flexible. The engine works by scanning your codebase and building a compact in-memory index that records sequences of three characters at a time. This approach allows it to index over ten million lines of code per minute. When you search for something, it looks up these sequences rather than scanning every file, which is why most queries return results in under a tenth of a second. You can search using simple text, regular expressions, and logical operators like AND, OR, and NOT, and you can filter by file name, repository name, or programming language. This tool is built for engineering organizations dealing with scale. For example, a platform team managing a large microservices architecture could use it to quickly locate every instance of a deprecated API across all services before a migration. A security team could search for vulnerable code patterns across the entire company codebase in real time. It supports access permission controls, so different teams or contractors only see the code they are authorized to view. The project provides two ways to interact with it: a web-based search interface and a command-line tool for developers who prefer working in the terminal. It also integrates with ctags, which means it can recognize and search for code symbols like variables, functions, and class names rather than just matching raw text. The project is open source under the Apache License, making it free for commercial use.