FakeGit 行动自 2025 年 3 月起活跃,利用近 7600 个恶意 GitHub 仓库(其中超过 800 个伪装成 AI 技能或 MCP 服务器)诱骗 Claude Code、Gemini CLI、ChatGPT 等 AI 编程助手自主发现并推荐恶意软件。
该行动采用名为 AgentBaiting 的新型攻击技术,通过虚假的 AI 技能注册表条目和精心伪造的 README 文件,将 AI 助手变为恶意软件分发链中的一环。最终载荷为信息窃取型恶意软件 StealC。
截至 2026 年 7 月,攻击者控制了约 6600 个 GitHub 账号,在公开 AI 注册表中发布了超过 600 条恶意清单,相关文件的 GitHub 下载计数事件超过 1400 万次。研究人员建议企业建立经批准的 AI 组件目录并限制 AI 助手的自主行为。
Search & fact-check with cited sources for What was the FakeGit campaign, how did it use nearly 7,600 fake GitHub repositories to trick AI cThe AgentBaiting technique weaponizes AI coding agents' trust, turning them into unwitting recommenders of malware-laden repositories.
AI 提示
Create a landscape editorial hero image for this Studio Global article: Search & fact-check with cited sources for What was the FakeGit campaign, how did it use nearly 7,600 fake GitHub repositories to trick AI c. Article summary: Here is a comprehensive, source-cited breakdown of the FakeGit campaign.. Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clickbait thumbnails, icons, and tiny thumbnail layouts. Make it useful as an illustrative visual, not as factual evidence.
FakeGit 行动自 2025 年 3 月起活跃,利用近 7600 个恶意 GitHub 仓库(其中超过 800 个伪装成 AI 技能或 MCP 服务器)诱骗 Claude Code、Gemini CLI、ChatGPT 等 AI 编程助手自主发现并推荐恶意软件。
该行动采用名为 AgentBaiting 的新型攻击技术,通过虚假的 AI 技能注册表条目和精心伪造的 README 文件,将 AI 助手变为恶意软件分发链中的一环。最终载荷为信息窃取型恶意软件 StealC。
截至 2026 年 7 月,攻击者控制了约 6600 个 GitHub 账号,在公开 AI 注册表中发布了超过 600 条恶意清单,相关文件的 GitHub 下载计数事件超过 1400 万次。研究人员建议企业建立经批准的 AI 组件目录并限制 AI 助手的自主行为。
Search & fact-check with cited sources for What was the FakeGit campaign, how did it use nearly 7,600 fake GitHub repositories to trick AI cThe AgentBaiting technique weaponizes AI coding agents' trust, turning them into unwitting recommenders of malware-laden repositories.
AI 提示
Create a landscape editorial hero image for this Studio Global article: Search & fact-check with cited sources for What was the FakeGit campaign, how did it use nearly 7,600 fake GitHub repositories to trick AI c. Article summary: Here is a comprehensive, source-cited breakdown of the FakeGit campaign.. Topic tags: general, general web, user generated. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clickbait thumbnails, icons, and tiny thumbnail layouts. Make it useful as an illustrative visual, not as factual evidence.