Anthropic Unveils Free AI-Powered Vulnerability Scanner for Open‑Source Projects

Anthropic announced on Thursday the launch of OSS Scanner, a free, AI‑driven tool that allows open‑source maintainers to run automated vulnerability checks on their codebases.

The service is optional and leverages the company’s Claude language model, drawing on insights gained while using the same technology to locate flaws during the internal Project Glasswing initiative.

Security of open‑source software has become a focal point for developers and enterprises after a series of high‑profile supply‑chain attacks highlighted how a single compromised library can affect thousands of downstream projects.

OSS Scanner integrates with popular repository platforms, enabling contributors to trigger scans via pull‑request comments or CI pipelines. While the tool highlights potential weaknesses, Anthropic emphasizes that it is intended to complement, not replace, traditional code review and testing practices.

Industry observers see the move as a sign that AI is maturing into a practical aid for software security. Anthropic plans to expand the scanner’s capabilities over time, inviting community feedback to refine detection rules and broaden language support.