好的，我现在需要帮用户总结这篇文章的内容，控制在100字以内。首先，我得仔细阅读文章，理解其主要内容。 文章主要讲的是Anthropic公司推出了一个新的AI工具，叫做Claude Code Security。这个工具用于扫描代码中的漏洞，并建议修复方法。它不同于传统的基于规则的扫描器，而是像人类研究员一样分析代码，关注数据流和组件交互，从而发现传统方法常常忽略的问题。 接下来，文章提到这个工具目前处于有限的研究预览阶段，面向企业、团队和开源维护者。Anthropic公司表示，这个工具经过多阶段验证，减少了误报，并且每个发现都有严重性和信心评分，帮助团队优先处理最重要的修复。 此外，Claude Code Security基于一年多的研究成果，在Capture-the-Flag活动中测试过，并与太平洋西北国家实验室合作开发。使用Claude Opus 4.6版本的团队已经在开源项目中发现了500多个长期未被发现的漏洞。 最后，Anthropic认为这是一个网络安全的关键时刻，AI将在未来扫描全球大部分代码，帮助防御者更快地发现和修复漏洞。 现在我需要将这些信息浓缩到100字以内。重点包括：工具名称、功能、优势（如分析方式）、验证过程、适用对象以及其重要性。 可能的结构是：Anthropic推出Claude Code Security AI工具，用于扫描代码漏洞并建议修复。该工具通过分析数据流和组件交互发现传统方法难以察觉的问题，并经过多阶段验证以减少误报。目前面向企业、团队和开源维护者开放预览。 </think> Anthropic推出Claude Code Security AI工具，用于扫描代码漏洞并建议修复。该工具通过分析数据流和组件交互发现传统方法难以察觉的问题，并经过多阶段验证以减少误报。目前面向企业、团队和开源维护者开放预览。 2026-2-23 00:11:55 Author: securityaffairs.com(查看原文) 阅读量:12 收藏
Anthropic unveils Claude Code Security to detect and fix code bugs
Pierluigi Paganini
February 23, 2026
Anthropic launches Claude Code Security, an AI tool that scans code for vulnerabilities and suggests how to address them.
Anthropic has introduced Claude Code Security, a new AI-powered service designed to scan software codebases for vulnerabilities and recommend fixes. Built into Claude Code, the tool aims to help teams detect and remediate security flaws faster. The capability is currently being rolled out in a limited research preview for Enterprise and Team customers.
“Claude Code Security, a new capability built into Claude Code on the web, is now available in a limited research preview. It scans codebases for security vulnerabilities and suggests targeted software patches for human review, allowing teams to find and fix security issues that traditional methods often miss.” reads the announcement published by Anthropic.
Claude Code Security is designed to discover complex, high-severity software vulnerabilities. Unlike traditional rule-based scanners, it analyzes code like a human researcher, focusing on data flows, understanding component interactions, and catching issues others miss. Each finding undergoes multi-stage verification, is assigned a severity and confidence rating, and appears in a dashboard for human review.
“Every finding goes through a multi-stage verification process before it reaches an analyst. Claude re-examines each result, attempting to prove or disprove its own findings and filter out false positives.” continues the announcement. “Findings are also assigned severity ratings so teams can focus on the most important fixes first.”
Currently in limited research preview for Enterprise, Team, and open-source maintainers, it empowers defenders while ensuring responsible deployment.
Claude Code Security leverages over a year of research into Claude’s cybersecurity skills. Tested in Capture-the-Flag events and in partnership with Pacific Northwest National Laboratory, Claude has been refined to detect and patch real vulnerabilities. Using Claude Opus 4.6, the team discovered 500+ long-undetected bugs in open-source projects. Now, Claude Code Security brings these defensive capabilities to Enterprise, Team, and open-source maintainers, letting teams review findings and iterate on fixes directly within familiar tools.
“This is a pivotal time for cybersecurity. We expect that a significant share of the world’s code will be scanned by AI in the near future, given how effective models have become at finding long-hidden bugs and security issues.” concludes Anthropic. “Attackers will use AI to find exploitable weaknesses faster than ever. But defenders who move quickly can find those same weaknesses, patch them, and reduce the risk of an attack.”
Anthropic is launching a limited research preview of Claude Code Security for Enterprise and Team customers, with early access and collaboration. Open-source maintainers can apply for free, expedited access.
Follow me on Twitter: @securityaffairs and Facebook and Mastodon
(SecurityAffairs – hacking, Claude Code Security)
如有侵权请联系:admin#unsafe.sh