Anthropic 的 Claude Security 随代码变更自动重扫仓库
Anthropic’s Claude Security now rescans repositories as code changes and marks findings fixed once patches land.
Anthropic 的 Claude Security 现在会持续监控仓库，代码变更落地后自动重扫，补丁上线后自动关闭对应问题，漏洞复现时重新打开，并结合安全团队笔记和可编辑威胁模型判断攻击者实际可达的漏洞。此前它只是按需或按计划运行的扫描器，代码一变结果就过期。Anthropic 称自 4 月以来已有数千家组织发现超过 100 万个漏洞；Claude Enterprise 客户可获得公测资格，以及截至 2026 年 12 月 31 日的两个月每月 2.5 万美元额度。
推荐理由
AI 评分 66
原文说明了 Claude Security 从按需扫描转为持续跟踪代码变化，读者可据此判断安全审查流程会怎样变化。
主题
查看原帖Claude Security 现在会随着代码的变更而保持同步。 只需连接一次你的仓库。Claude 会先执行一次深度扫描，随后在变更落地时重新扫描，因此发现结果始终贴近你代码的当前状态。 现已在 Claude Enterprise 上开启公开测试。https://t.co/J5kPSEhWp3
查看原文
Claude Security now keeps up with your code as it changes. Connect your repositories once. Claude runs a deep first scan, then re-scans as changes land, so findings stay close to the current state of your code. In public beta on Claude Enterprise. https://t.co/J5kPSEhWp3
正文
当前显示原文，中文全文译文尚未就绪。
Anthropic’s Claude Security now rescans repositories as code changes and marks findings fixed once patches land.
Until now, Claude Security was a scanner you ran (on demand or on a schedule) that handed you a list of bugs, and that list went stale as soon as your code changed.
But now it watches your repositories all the time, rescans each change as it lands, closes a finding by itself once the fix ships (wherever it was made), reopens it if the bug comes back, and uses notes from your security staff plus an editable threat model to judge which bugs an attacker could actually reach.
Anthropic says thousands of organizations have found over 1M vulnerabilities since April.
Claude Enterprise customers get the public beta plus $25K monthly credits for 2 months through December 31, 2026.