Security
Code Chaos: The Unintentional NPM Leak
Anthropic recently faced a situation where the source code for Claude Code, a closed-source platform, was accidentally leaked. Fortunately, no customer data or sensitive information was compromised in the process.
Although Anthropic has been a supporter of the open-source community, Claude Code has always been kept under wraps. However, an update inadvertently exposed the internal source code, leading to its dissemination across platforms like GitHub.
Anthropic, in a statement to BleepingComputer, acknowledged the leak and reassured that no personal or sensitive information was disclosed. They attributed the incident to a packaging error during a release and emphasized it was not a security breach.
The leaked source code, initially spotted by Chaofan Shou (@Fried_rice), contained details about Claude-exclusive features and approximately 1,900 files with 500,000 lines of code.
One of the key reasons for the exposure was the inclusion of a large .map file, which, if containing the “sourcesContent” field, can potentially reveal the entire source code tree when shared publicly.
Following the leak, Anthropic took steps to remove the source code from various platforms by issuing DMCA infringement notifications.
Developers quickly seized the opportunity to analyze the source code for undocumented features, uncovering functionalities like “Proactive mode” and “Dream” mode within Claude Code.
Notably, users reported a bug in Claude Code that led to a reduction in usage limits, causing concerns among customers. Anthropic acknowledged the issue and assured users that they were actively investigating the matter.
As of the latest update, Anthropic’s team continues to work on resolving the bug, which has affected users across different pricing plans.
Some users speculate that the changes in usage limits could be intentional, considering Claude’s increasing popularity in recent weeks. However, further details from Anthropic are necessary to confirm this speculation.

Source: BleepingComputer
Anthropic Investigates Claude Code Usage Bug
Another issue that surfaced was the alleged reduction in usage limits for Claude Code users, prompting Anthropic to address the concerns raised by customers.
Users reported unexpected behavior in usage limits, with some experiencing faster depletion than anticipated. Anthropic acknowledged the issue and assured users of ongoing investigations to resolve the bug.
Despite efforts to rectify the situation, the bug remains unresolved as of the latest update, emphasizing the significance of the matter to Anthropic’s team.
Speculations surrounding the intentional nature of the changes in usage limits have emerged, considering Claude’s popularity surge. However, conclusive details from Anthropic are awaited to clarify the situation.
Automated pentesting proves the path exists. BAS proves whether your controls stop it. Most teams run one without the other.
This whitepaper maps six validation surfaces, shows where coverage ends, and provides practitioners with three diagnostic questions for any tool evaluation.
-
Facebook5 months agoEU Takes Action Against Instagram and Facebook for Violating Illegal Content Rules
-
Facebook6 months agoWarning: Facebook Creators Face Monetization Loss for Stealing and Reposting Videos
-
Facebook6 months agoFacebook Compliance: ICE-tracking Page Removed After US Government Intervention
-
Facebook4 months agoFacebook’s New Look: A Blend of Instagram’s Style
-
Facebook4 months agoFacebook and Instagram to Reduce Personalized Ads for European Users
-
Facebook6 months agoInstaDub: Meta’s AI Translation Tool for Instagram Videos
-
Facebook4 months agoReclaim Your Account: Facebook and Instagram Launch New Hub for Account Recovery
-
Apple5 months agoMeta discontinues Messenger apps for Windows and macOS

