用Zig打造的高性能命令卫士，本地守护编码代理，86个安全包拦截危险操作，MCP调用本地分类，会话不离开电脑。
Article URL: https://github.com/christopherkarani/Orca Comments URL: https://news.ycombinator.com/item?id=48782806 Points: 2 # Comments: 0
为MCP代理加一道安全闸：轻量级策略执行层，一行命令即可部署，想给AI Agent立规矩的开发者值得一试。
Article URL: https://github.com/Dnakitare/imara Comments URL: https://news.ycombinator.com/item?id=48663284 Points: 1 # Comments: 0
AI代理数据安全新框架，声明式策略执行保障数据流控制
arXiv:2606.05679v1 Announce Type: cross Abstract: Agents increasingly generate SQL, orchestrate pipelines, and automate data analysis on behalf of use…
提出 ConsisGuard 方法，对齐安全审议与策略执行，提升大模型护栏的可靠性与一致性。
arXiv:2605.31073v1 Announce Type: new Abstract: Reasoning-based LLM guardrails improve safety moderation by generating explicit rationales before issu…