Grimlock: 使用eBPF和认证通道保护高代理系统
Grimlock: Guarding High-Agency Systems with eBPF and Attested Channels
AI总结:
针对代理系统中用户编排代码带来的安全挑战,提出Grimlock代理守卫,通过eBPF强制流量拦截和TLS 1.3通道绑定认证,实现透明、可审计、作用域绑定的代理间通信。
摘要:
代理系统越来越多地运行用户编写的编排代码,这些代码调用工具、生成子任务并在机器和云之间委派工作。虽然这种高代理效率很高,但它带来了安全问题:身份、授权、来源和委派通常被推入应用程序代码,在那里它们变得难以一致地执行和审计。我们提出Grimlock,一种代理守卫,通过将信任执行移动到沙箱子系统中,同时保持代理代码不变,来恢复关注点分离。Grimlock使用eBPF强制流量拦截来确保沙箱通信通过守卫,并将其与绑定到标准TLS 1.3通道绑定的握手后认证相结合。通道建立后,守卫授权通信并生成短暂的、通道绑定的作用域令牌,这些令牌捕获最小权限委派。在接收端,目标守卫重新验证身份、作用域和通道绑定,终止TLS,并仅在策略检查成功后向目标沙箱释放明文。kTLS为受保护的通信提供了高效的数据平面。因此,Grimlock提供了一条路径,使用通用Linux原语,无需更改用户层编排代码,即可在异构多云环境中实现透明、可审计、作用域绑定的代理间通信。
英文摘要:
Agentic systems increasingly run user-authored orchestration code that invokes tools, spawns subtasks, and delegates work across machines and clouds. Although this high agency is productive, it creates a security problem: identity, authorization, provenance, and delegation are often pushed into application code, where they become difficult to enforce consistently and difficult to audit. We present Grimlock, an Agent Guard that restores separation of concerns by moving trust enforcement into the sandbox substrate while leaving agent code unchanged. Grimlock uses eBPF-enforced traffic interception to ensure that sandbox communication passes through a guard, and combines it with post-handshake attestation bound to standard TLS~1.3 channel bindings. After a channel is established, the guard authorizes communication and mints short-lived, channel-bound scope tokens that capture least-privilege delegation. At the receiving side, the destination guard re-validates identity, scope, and channel binding, terminates TLS, and releases plaintext to the destination sandbox only after policy checks succeed. kTLS provides an efficient dataplane for protected communication. As a result, Grimlock offers a path toward transparent, auditable, and scope-bound agent-to-agent communication across heterogeneous multi-cloud environments, using commodity Linux primitives and without requiring changes to user-layer orchestration code.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。