Anthropic has rolled out new security tools for its AI assistant, Claude Code. These updates are designed to help developers catch security issues early in the coding process. The company hopes this will make their tool stand out in a crowded field of AI coding assistants.

Making Security Checks Easier for Developers

One of the biggest new features is a command called “/security-review”. Developers can run this in Claude Code to scan their code for vulnerabilities. It’s like having a security expert look over your shoulder. The command checks for common problems like SQL injection, cross-site scripting, and problems with authentication. It then explains what it found and suggests fixes.

This makes it easier for developers to find and fix security issues before they get into production. Instead of waiting for manual reviews, they can run quick checks themselves. The idea is to catch problems early, when they’re easier to fix and less costly.

Better Integration with Development Workflows

Another new feature is a GitHub Action that works with Claude Code. It automatically scans code whenever a pull request is opened. It looks for vulnerabilities, applies rules to reduce false alarms, and then posts comments directly in the pull request. This helps teams keep security checks consistent across all code changes.

The goal is to integrate security into the regular development process. Instead of separate security reviews, teams can now have security checks happen automatically during their usual workflow. This could help prevent insecure code from reaching production and make security an ongoing part of development.

Why This Matters in the Growing AI Coding Space

Anthropic’s move comes at a time when AI tools are becoming more popular among developers. A 2025 survey found that 84% of developers are using or plan to use AI in their work, up from 76% the year before. But trust is still an issue. Only a third of developers fully trust AI-generated code, and many remain cautious about its accuracy.

With these new security features, Anthropic is trying to show that Claude can not only help write code faster but also make it safer. Experts say that these tools could help teams work more securely without adding too much extra work. By automating the security review process, teams can focus on building features while still maintaining good security practices.

What the Future Might Hold for AI in Software Security

The addition of automated security reviews marks a shift in how AI tools are used in software development. Instead of just helping write code, AI is now taking on a role in security and governance. This could lead to more reliable and secure software, especially as AI tools become more advanced.

However, experts remind us that AI is not perfect. While Claude’s large context window helps it understand complex code, it can still produce false positives or miss issues. Companies using these tools need to keep manual review processes in place and treat AI suggestions as helpful hints, not foolproof answers.

Overall, Anthropic’s new features aim to make AI a more responsible partner in development. If successful, they could push the industry toward more secure coding practices, while also raising questions about how much trust we should place in AI-driven security checks. As the AI coding landscape heats up with upcoming releases like GPT-5 and other big players recruiting top talent, the race to build smarter, safer tools is only getting more intense.

Inspired by

• https://www.infoworld.com/article/4035583/anthropic-targets-devsecops-with-claude-code-update-as-ai-rivals-gear-up.html

Sources

• gmpg.org
• stackoverflow.co
• csoonline.com
• csoonline.com
• greyhoundresearch.com