Anthropic Launches Opt-in Vulnerability Scanner for Open-Source Software

Anthropic Launches Opt-in Vulnerability Scanner for Open-Source Software Anthropic has introduced OSS Scanner, a free, opt-in service using its most capable models—including Claude Mythos—to conduct periodic automated security scans and provide vulnerability reports to eligible open-source projects. Over six months, Anthropic identified more than 29,000 candidate vulnerabilities in major software projects but faced a bottleneck, manually triaging only about 6,000. Nearly 5,000 unvalidated reports with proposed patches have already been sent directly to maintainers who requested bulk submissions. OSS Scanner reports are fully model-generated without human triage, bundling reproducers, vulnerability explanations, and candidate patches where available. Testing 97 critical and high-severity findings across 48 projects showed 85 (88%) met coordinated vulnerability disclosure standards, with only one false positive. Scaling Beyond Human Triage Bottlenecks Recent advances have increased LLM vulnerability detection rates on the CyberGym benchmark from under 20% early last year to over 85% this year. Over the past six months, Anthropic detected over 29,000 candidate vulnerabilities in critical software but could only manually review approximately 6,000. Because exploits can now be built in minutes, maintainers increasingly requested bulk, unverified reports to remediate issues ahead of attackers, prompting the creation of a fast-track automated scanner alongside the standard coordinated disclosure process. OSS Scanner Mechanics and Model Automation Modeled after Google's OSS-Fuzz, OSS Scanner provides security audits free of charge to open-source software using Anthropic's strongest models, such as Claude Mythos. Findings bypass human review to allow faster, more frequent scans, meaning some reports may be invalid or incorrect. Each generated report includes a self-contained reproducer, a bug explanation with bisection data when possible, and a suggested fix patch. Validation Accuracy and Project Eligibility In a validation test of 97 high and critical severity findings across 48 projects, penetration testers found that 85 met disclosure standards, 11 were real but duplicate issues, and only one was an outright false positive. While some maintainers noted occasional inflated severity ratings or threat model misunderstandings, invalid reports remained rare. Critical infrastructure open-source projects can enroll on a case-by-case basis by opening a pull request in Anthropic's GitHub repository.

Anthropic 推出开源软件自选漏洞扫描服务 OSS Scanner Anthropic 宣布推出开源项目自选漏洞扫描服务 OSS Scanner，利用最强模型定期提供免费安全扫描，以全自动生成报告的方式加快漏洞发现与修复速度。 在过去六个月中，模型发现逾 29,000 个候选漏洞，但人工仅能审核约 6,000 个，导致披露存在人力瓶颈。 OSS Scanner 产出完全由模型生成且未经人工分流，附带独立复现步骤、漏洞说明与修复补丁建议。 早期测试中，专家抽检 97 个严重/高危漏洞，88% 达到 CVD 标准，仅 1 个为误报。 符合对基础设施和用户安全有关键影响等标准的开源项目，可通过提交 GitHub PR 申请加入。 人工审核瓶颈与快速披露通道 大语言模型在漏洞发现基准 CyberGym 上的检出率从去年初的不足 20% 攀升至今年的 85% 以上。过去六个月，Anthropic 模型在重点软件中识别出超 29,000 个候选漏洞，但受限于人力仅人工审核约 6,000 个。由于当前攻击者数分钟内即可开发漏洞利用，已有维护者主动要求接收未经审核的批量报告。为此，Anthropic 在保留协调漏洞披露（CVD）流程的同时，推出该自选快速通道。 OSS Scanner 机制与早期测试表现 灵感源自谷歌 OSS-Fuzz，OSS Scanner 采用包括 Claude Mythos 在内的顶尖模型，向开源项目提供免费且全自动的代码安全审计。虽然未经人工分流可能带来无效报告，但每份报告均包含独立复现用例、漏洞解释及候选补丁。在对 48 个项目的 97 个严重与高危漏洞抽样中，85 个符合 CVD 披露标准，11 个属已知或重复问题，仅 1 个误报；维护者反馈存在个别危害定级偏高或威胁模型理解偏差的情况。 项目申请资格与配套支持 核心维护者可通过在指定 GitHub 仓库提交 PR 申请加入。入选标准参考 OSS-Fuzz，重点关注对基础设施和用户安全具有关键影响的项目。此外，Anthropic 还通过网络验证计划为合格安全专家开放进阶功能，并经由 Claude for OSS 提供免费的 Claude Max 20x 订阅支持开源漏洞修复。

*We’re launching [OSS Scanner](https://red.anthropic.com/oss-scanner), an opt-in vulnerability scanner for the open-source ecosystem informed by our experience using Claude to find vulnerabilities during [Project Glasswing](https://www.anthropic.com/glasswing). Projects that join will receive thorough, periodic security scans by our strongest models at no cost.* Language models have rapidly advanced in their ability to [discover vulnerabilities](https://www.anthropic.com/research/mythos-preview), as we have [reported extensively](https://www.anthropic.com/research/zero-days) on this blog. On [CyberGym](https://arxiv.org/abs/2506.02548), one academic vulnerability-finding benchmark, LLMs have gone from finding under 20% of vulnerabilities at the beginning of last year to finding over 85% this year. As a result, open source maintainers have gone from receiving [mostly slop](https://daniel.haxx.se/blog/2025/07/14/death-by-a-thousand-slops/) from LLMs to receiving [high-quality bug reports](https://lore.kernel.org/linux-media/CAHk-=wi4zC+Ze8e+p3tMv8TtG_80KzsZ1syL9anBtmEh5Z40vg@mail.gmail.com/). Over the last six months, we’ve used our latest models to scan for vulnerabilities in some of the world’s most important software projects. We have discovered [over 29,000 candidate vulnerabilities](https://red.anthropic.com/2026/cvd/), but have only been able to manually review and triage approximately 6,000 of these. While 6,000 vulnerabilities is significant, we remain bottlenecked on our human capacity to validate these findings. We are working on scaling our vulnerability disclosures. Meanwhile, with increasing frequency, maintainers who receive our first reports simply ask us for a bulk submission of all the unverified reports with proposed patches: to date, we have sent nearly 5,000 reports directly to maintainers after they asked to receive everything we had—even if it wasn’t validated. Since exploits can now be developed [in minutes](https://www.anthropic.com/research/n-days), projects that can find and address vulnerabilities faster can better secure their software against attackers who are racing to find and leverage these same weaknesses. We will continue to manually disclose human-verified vulnerability reports via our existing [coordinated vulnerability disclosure (CVD) process](https://www.anthropic.com/coordinated-vulnerability-disclosure), especially for projects without the resourcing to triage reports themselves. But we are now making available an optional fast-track for those who wish to receive vulnerability reports as soon as they become available. Inspired by the positive impact on the open source ecosystem of [Google's OSS-Fuzz](https://google.github.io/oss-fuzz/), a project that scans open-source software for vulnerabilities with fuzzers, we are launching [OSS Scanner](https://red.anthropic.com/oss-scanner) to find vulnerabilities in open-source code with our strongest language models. While [Claude Security](https://claude.com/product/claude-security), our general-access code scanning and patching product focuses on helping enterprises defend their systems, OSS Scanner provides security audits at no cost to open-source projects. The outputs of this opt-in vulnerability scanner will be fully model-generated, without human review or triage. This will enable faster and more frequent scanning, but means that it is possible reports will be incorrect or invalid. These reports will be generated by our strongest models (including Claude Mythos) to give open-source projects the largest defensive advantage. We have spent the last several weeks validating this pipeline with dozens of open source projects. These initial disclosures contained hundreds of bug reports, including multiple vulnerabilities that we were able to chain to unauthenticated remote code execution exploits affecting these projects. Each report contains a self-contained reproducer, explanation of the vulnerability (including a bisection to determine when the bug was introduced, where possible), and candidate patch (when available) for how the bug can be fixed. We are now making this service available to more open source projects. Some of the feedback we received as we tested early versions of OSS Scanner and our raw model outputs: To validate an early version of OSS Scanner, we asked the expert penetration testers who review our CVD findings to check 97 critical and high-severity vulnerabilities from the scanner across 48 projects. Of these, 85 (88%) met the bar for our CVD process. Of the remaining 12, 11 were real but duplicated known issues or other findings from the scan, and only one was invalid, i.e. a “false positive.” We’ve since sent many of these findings to maintainers, who have seldom told us a high or critical finding was invalid, consistent with [our prior open-source work](https://red.anthropic.com/2026/cvd/). Some have told us severity ratings can be inflated or the scanner misunderstood the project’s threat model. We can’t guarantee the scanner will be perfect, but we’ll keep refining the system based on maintainer feedback and as models improve. Core maintainers of eligible projects can enroll by submitting a PR to [this GitHub repo](https://github.com/anthropics/oss-scanner) following the standard project template, with additional guidance provided in [our extended FAQ](https://red.anthropic.com/oss-scanner). Projects are eligible based on a similar set of criteria that OSS-Fuzz uses: briefly, projects should have a “critical impact on infrastructure and user security” and we will make decisions on a case-by-case basis. *Whether or not OSS Scanner is right for you, our new [Cyber Verification Program](https://www.anthropic.com/news/cyber-verification-program) makes advanced cyber capabilities and reduced blocking classifiers available to qualifying security professionals. [Claude for OSS](https://claude.com/contact-sales/claude-for-oss) provides free Claude Max 20x subscriptions to help remediate vulnerabilities and improve OSS projects.*