Tag
svpino 推介开源工具 XERJ Corpus Hub，它可作为代理的"参考记忆"引擎接入任意 AI 模型，通过本地索引的代码、文档与安全知识库进行检索，让代理无需再训练即可提升表现。
The user is attempting to reproduce Booz Allen's claims about the vulnerability of Chinese AI models, reporting that Qwen 3 has produced the best results so far, while questioning the report's financial motivations, reproducibility, and lack of peer review.
Cognition introduces Devin Security Swarm, a new tool for finding security vulnerabilities using an Agentic MapReduce architecture, achieving higher accuracy and lower cost than alternatives.
This post discusses Devin's new 'Security Swarm' feature using Agentic MapReduce to scale AI-driven code security analysis, illustrating the need for 100× more AI inference and the strategic deployment of diverse models across industries.
Anthropic released an open-source code auditing reference harness for autonomous vulnerability discovery and remediation using Claude, covering a recon→find→triage→report→patch pipeline, primarily targeting C/C++ memory vulnerabilities. It is a template/reference implementation rather than a production-ready product, with a managed hosted option called Claude Security also available.
Google is expanding its CodeMender AI agent for code security to compete with Anthropic's Mythos, targeting governments and enterprises as AI labs see cybersecurity as a key revenue driver.
Google DeepMind introduces CodeMender, an AI agent that automatically detects and fixes code security vulnerabilities using advanced reasoning and validation techniques. The system has already upstreamed 72 security fixes to open source projects over six months.
Block shared how it uses OpenAI’s AI capabilities to discover and fix software security risks—the tool goes beyond statically scanning code to reason about attack patterns, cross trust boundaries, and chain together multistep attacks, revealing complete, end-to-end compromise paths while helping pinpoint issues, identify owners, suggest patches, and verify that fixes are effective.