CASE EVIDENCE / A RECORD
CyberStrikeus 公开的代码审查与测试案例，来源为 公开代码库，复核于 2026-06-26T17:19:07Z。
原始记录：CyberStrike — Autonomous Penetration Testing Agent with DeepSeek V3 Backend
Chinese Brief
CyberStrikeus 公开的代码审查与测试案例，来源为 公开代码库，复核于 2026-06-26T17:19:07Z。 Model Atlas 将它标记为 A 类证据，因为它同时具备具体使用者、具体任务、公开原始证据和可访问产物。 Model Atlas 不把 benchmark、教程、发布说明或集合页包装成真实案例。
任务
这是一个围绕代码审查和测试生成的真实任务，公开材料可以回溯到具体使用者和具体产物。 原始资料写作：Integrated DeepSeek V3 as a supported LLM backend for CyberStrike, an AI-powered offensive security agent with 7,300+ actionable security skills based on MITRE ATT&CK. DeepSeek V3 is positioned as a cost-effective alter…
公开材料提供公开代码、README 或项目配置，可用于核验任务结果、项目形态和模型绑定关系。 原始资料写作：An open-source AI pentesting agent with 7,300+ security skills supporting multiple LLM backends including DeepSeek V3. The platform enables autonomous penetration testing workflows where DeepSeek V3 provides the reasoni…
DeepSeek V3 (Dec) 在该案例中承担代码审查和测试生成相关的生成、分析、编排或实现角色。 原始资料写作：DeepSeek V3 serves as one of the primary LLM backends for CyberStrike's autonomous pentesting capabilities, specifically positioned for cost-effective operation. It processes security assessment tasks including vulnerab…
当前判断基于公开材料；若产物下线、仓库变更或模型参与比例仅来自作者自述，需要在引用前重新复核。 原始资料写作：DeepSeek V3 is one of many supported LLM backends, not the exclusive model. Security tooling use case. 716 stars indicates emerging rather than established adoption.