Anthropic Updates Claude AI Code With Security Plugin

Key Takeaways

• Anthropic has released a new real-time security-guidance plugin for its Claude Code AI terminal coding tool.
• The plugin identifies approximately 25 high-risk vulnerability classes, including SQL injection and XSS, directly within the development workflow.
• The update also brings significant performance enhancements, clearer error messages, and increased rate limits for users, bolstered by a new infrastructure deal with SpaceX.

Anthropic Enhances Claude Code with Real-Time Security Plugin and Performance Boosts

Anthropic has rolled out a significant update for Claude Code, its AI-powered terminal coding environment, introducing a new real-time security-guidance plugin. This enhancement aims to integrate immediate vulnerability detection directly into the developer’s workflow, alongside broader performance improvements designed to streamline the coding experience.

Real-Time Vulnerability Detection Integrated

The newly available security-guidance plugin can be activated by any Claude Code user through the /plugins command within the marketplace. Once installed, it actively monitors code edits, diffs, and commits, flagging dangerous patterns before they can progress to production environments.

Operating on regex-based pattern matching, the plugin is capable of identifying around 25 categories of high-risk vulnerabilities. These include critical issues such as SQL injection, command injection, cross-site scripting (XSS), hardcoded API keys and secrets, insecure deserialization, and improper input validation—flaws frequently exploited in real-world breaches.

We’ve shipped a security-guidance plugin for Claude Code that helps identify and fix vulnerabilities as you’re writing code.

Available for all Claude Code users. Install from the plugin marketplace (/plugins). pic.twitter.com/LprgC4m6Kf

— ClaudeDevs (@ClaudeDevs) May 26, 2026

When a potential security risk is detected, Claude Code provides inline suggestions for fixes during the same coding session, eliminating the need for developers to switch to external security scanning tools. This approach is complemented by Anthropic’s broader Claude Security platform, which offers more in-depth, AI-driven codebase scanning, traces data flows, performs adversarial verification, and proposes targeted patches for human review.

Internal metrics from Anthropic reportedly indicate a 30–40% decrease in security-related comments on pull requests since the introduction of this tool. The announcement, shared via the official Claude Developers account on X (formerly Twitter), confirms the plugin’s immediate availability across all user tiers.

Performance Enhancements and Infrastructure Expansion

Beyond security, Anthropic has addressed several user experience issues reported by the developer community. Key improvements include:

• Full-screen renderer: A new renderer aims to eliminate screen flickering that can disrupt prolonged coding sessions.
• Faster streamed responses: Reduced latency for more fluid code generation and editing.
• Clearer error messages: Enhanced feedback for developers when issues arise within Claude Code.
• Eased rate limits: Pro and higher plan users now benefit from doubled session caps, made possible by a new compute agreement.

These performance upgrades are underpinned by a landmark infrastructure deal with SpaceX, granting Anthropic access to the entire compute capacity of the Colossus 1 data center in Memphis, Tennessee. This agreement is set to provide over 300 megawatts of new capacity, equivalent to more than 220,000 NVIDIA GPUs, which will come online within weeks.

First for our new full-screen renderer (which should get rid of bugs like screen flickering), we’ve made a number of fixes for different environments and terminals.

You can turn it with the command: /tui feedback

We’re working on making it the default in Claude Code soon.

— ClaudeDevs (@ClaudeDevs) May 27, 2026

Initial feedback from the developer community has been largely positive, with users noting a more fluid UI and seamless integration of the security plugin without workflow interruptions. While some minor UI quirks persist, Anthropic has acknowledged ongoing refinement efforts, positioning these updates as part of a continuous strategy to enhance Claude Code’s responsiveness and reliability in a competitive AI coding assistant market, directly challenging tools like GitHub Copilot and Cursor.

The integration of automated vulnerability detection at the diff level represents a significant “shift left” in security, moving detection into the early coding phase rather than relying solely on later-stage static application security testing (SAST) tools or CI/CD pipeline checks. The security plugin is now accessible to all users via the /plugins command within Claude Code.

What You Should Do

• Activate the Security Plugin: If you are a Claude Code user, immediately install the new security-guidance plugin using the /plugins command in your terminal.
• Review Inline Fixes: Pay close attention to the inline fix suggestions provided by Claude Code when risky patterns are identified and integrate them into your development process.
• Stay Updated: Regularly check for further updates from Anthropic to ensure you benefit from the latest security and performance enhancements.
• Integrate with Broader Security Practices: While the plugin provides real-time feedback, it should complement, not replace, comprehensive security testing, code reviews, and SAST/DAST tools in your CI/CD pipeline.

Disclaimer: HackersRadar reports on cybersecurity threats and incidents for informational and awareness purposes only. We do not engage in hacking activities, data exfiltration, or the hosting or distribution of stolen or leaked information. All content is based on publicly available sources.