Anthropic Claude Code Source Code Leak Triggers Widespread GitHub Takedowns

A routine attempt to contain a software leak briefly spiraled into a broader disruption on GitHub this week, after Anthropic moved to remove copies of its own code from public repositories and inadvertently swept up thousands of unrelated projects in the process.

The issue began when a software engineer noticed that a recent release of Anthropic’s Claude Code command line tool appeared to include unintended access to its underlying source code. The discovery quickly circulated among developers and AI enthusiasts, with copies of the code shared across GitHub for closer inspection.

Anthropic responded by issuing a takedown request under U.S. copyright law, asking GitHub to remove repositories hosting the leaked material. According to GitHub’s own records, the enforcement action extended far beyond its apparent target, affecting around 8,100 repositories.

Among those caught in the sweep were legitimate forks of Anthropic’s publicly available Claude Code repository. Developers whose projects were temporarily blocked took to social media to express frustration, noting that their work had no connection to the leaked content.

The company later acknowledged the scope of the mistake. Boris Cherny, who leads the Claude Code project, said the takedown had been broader than intended due to how the targeted repository was linked within GitHub’s fork network.

In a statement shared with TechCrunch, an Anthropic spokesperson said the company had moved to correct the error after identifying the problem. The original notice was withdrawn in most cases, leaving only one repository and 96 associated forks subject to removal because they contained the accidentally released code.

GitHub subsequently restored access to the affected repositories once the revised request was processed.

The episode adds an awkward moment for Anthropic at a time when operational precision carries added weight. The company is reportedly preparing for a public offering, a process that often brings heightened scrutiny to internal controls and execution.

For developers affected by the temporary removals, the incident serves as a reminder of how quickly automated enforcement systems can scale beyond their intended scope, particularly when large code networks are involved.

Read More:

Oracle Layoffs Bring Personal Stories Into Focus as One Employee’s Response Resonates

Anthropic, for its part, has moved to contain both the leak and its unintended consequences, though the incident has already drawn attention within the developer community.