How we contain Claude across different products

我们如何对不同产品中的Claude进行隔离控制

Anthropic describes three mechanisms for containing Claude agent deployment risks across products: sandboxing or VMs, network egress controls, system-prompt and training constraints, and fine-grained permissions for MCP servers and third-party plugins.

Why it matters: Anthropic discloses a concrete containment stack for Claude agents, stronger than a routine product note. HKR-H/K/R all pass, but this is not a model launch or major capability release, so it stays in the 78–84 band.