Anthropic has announced Claude Code Security, a new capability built into Claude Code, now available in a limited research preview.

This feature scans codebases for security vulnerabilities and suggests targeted software patches for human review, enabling teams to find and fix security issues that traditional methods often miss.

Unlike static analysis, which typically scans for known patterns, Claude Code Security reads and reasons about code like a human security researcher would. It understands how components interact, traces data flow through the application, and catches complex vulnerabilities that rule-based tools might overlook.

Findings undergo a multi-stage verification process to filter out false positives. Claude re-examines each result, attempting to prove or disprove its own findings. Severity ratings are assigned so teams can focus on the most critical fixes first.

Validated findings appear in the Claude Code Security dashboard, where teams can review them, inspect suggested patches, and approve fixes. Claude also provides a confidence rating for each finding, accounting for nuances difficult to assess from source code alone. All actions require human approval.

Claude Code Security builds on over a year of research into Anthropic's Claude's cybersecurity capabilities. The company's Frontier Red Team has systematically tested these abilities through competitive events and collaborations, such as experimenting with AI to defend critical infrastructure.

Using Claude Opus 4.6, the team discovered over 500 vulnerabilities in production open-source codebases that had gone undetected for decades. Anthropic is working with maintainers on triage and responsible disclosure and plans to expand its security work with the open-source community.

Anthropic also uses Claude for its own code reviews, finding it highly effective for securing its systems. Claude Code Security aims to make these defensive capabilities more widely accessible, allowing teams to review findings and iterate on fixes within their existing tools.

The company anticipates that AI will scan a significant portion of the world's code in the near future due to its effectiveness in finding hidden bugs and security issues. While attackers will leverage AI to find exploitable weaknesses faster, defenders who adopt these tools can identify and patch those same weaknesses, thereby reducing the risk of attacks. Claude Code Security represents a step towards more secure codebases and a higher industry-wide security baseline.

👉 Read the full article on Anthropic

FitWit AI is an AI-powered fitness app that offers personalized workout plans that adapt to the user's progress.

Instead of following generic routines, the app creates workouts based on the user's goals, fitness level, and available equipment, then continuously improves them using workout data. Whether aiming to build muscle, lose fat, increase strength, or maintain a fitness routine, FitWit AI acts as an intelligent workout companion.

FitWit AI creates customized workout plans tailored to fitness goals (muscle gain, fat loss, strength, endurance), experience level (beginner to advanced), and available equipment (gym, dumbbells, or bodyweight). Each workout includes structured exercises with recommended sets and reps to help train effectively and safely.

As users log their workouts, FitWit AI analyzes their performance and automatically adjusts future workouts. This ensures training evolves as users get stronger, helping them avoid plateaus and stay challenged.

The app allows for easy logging of sets, reps, and weights, tracking of workout history and performance trends, and monitoring of progress over time to stay motivated and consistent. Everything is designed to be simple, clear, and easy to use.

FitWit AI is suitable for beginners seeking structured guidance, intermediate and advanced users wanting adaptive workouts, and for both gym training or home workouts.

Users set up their fitness profile and goals, choose their available equipment, and receive an AI-generated workout plan. As they log workouts and track progress, the AI adapts future workouts based on their performance.

The app promises smart, goal-focused training without complicated setup or generic plans.

👉 Read the full article on App Store