An error during the publication of the package led to the leakage of the internal code of the Claude Code tool, which allowed potential attackers to study its operation from the inside.
An error during the publication of the package led to the leakage of the internal code of the Claude Code tool, which allowed potential attackers to study its operation from the inside.
According to The Hacker News, approximately 500,000 lines of code and about 2,000 TypeScript files were exposed to the public. The cause was an incorrectly compiled npm package that left behind service files that allowed the full structure of the project to be reconstructed.
This is not about individual fragments, but actually about the complete logic of the tool. The leak contains code that is responsible for executing system commands, accessing the file system, processing requests, and interacting between internal components. This allows us to understand how Claude Code works at the architectural level, and not just through the interface.
The researchers also paid attention to the internal mechanisms of process launch. In particular, the code describes how the tool executes shell commands, manages subprocesses, and processes the results of their execution. Such details can be critical from a security point of view, as they allow for more precise selection of attack scenarios.
The leak also hints at features that haven’t been publicly revealed yet, including autonomous modes where the AI performs tasks in the background without constant user intervention. This effectively reveals some of the product’s future development.
The incident also poses practical risks for developers. Attackers could create malicious packages that mimic official tools or their dependencies. The article notes that such attempts have already been recorded in the npm ecosystem, where supply chain attacks are becoming increasingly common.
The incident comes amid a series of attacks in the JavaScript ecosystem, where malicious code is distributed through dependencies in popular packages. In such cases, installing or updating a library is enough to compromise a system, so internal code leaks only exacerbate these risks.
Previously, dev.ua wrote about how the popular JavaScript library Axios, used in thousands of projects, was compromised through a supply chain attack. In the infected versions, hackers hid a Trojan that could get into applications along with dependency updates.
