---
id: 20260605-T0-05
title: "Anthropic开源AI漏洞发现框架Defending Code"
title_en: "Anthropic Open Sources AI Vulnerability Discovery Framework"
url: https://ai.daily.yangsir.net/daily/20260605-T0-05
issue_date: 2026-06-05
publish_date: 2026-06-04T20:11:20.000Z
category: release
source_name: "HN AI 精选"
source_url: https://github.com/anthropics/defending-code-reference-harness
---

# Anthropic开源AI漏洞发现框架Defending Code

Anthropic发布了开源框架Defending Code，用于AI驱动的漏洞发现。该框架结合静态分析和动态测试，能自动识别代码中的潜在安全风险。项目在GitHub开源，包含80+评论，获得240+赞。初步测试显示，该框架在常见漏洞检测上比传统工具快3倍，准确率提升25%。

## English Version

**Anthropic Open Sources AI Vulnerability Discovery Framework**

Anthropic has open-sourced Defending Code, an AI-powered vulnerability discovery framework. The framework combines static analysis and dynamic testing to automatically identify potential security risks in code. Available on GitHub with 80+ comments and 240+ upvotes, early tests show it runs 3x faster than traditional tools with 25% higher accuracy in common vulnerability detection.

---

**来源**：[HN AI 精选](https://github.com/anthropics/defending-code-reference-harness)

**详情页**：https://ai.daily.yangsir.net/daily/20260605-T0-05

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*