Anthropic adds per-end-customer enforcement tracking and bans to the Claude API for platforms and resellers

Anthropic released API version 2026-10-21 with new client SDK majors, and added "user profiles" with enforcement_status and enforcement_actions fields so platforms, gateways, and resellers can see whether Anthropic has restricted or banned a specific end customer rather than their whole organization. It also added beta organization webhooks for restriction/ban events, took inference hooks out of beta for Enterprise orgs, and extended organization-level web fetch/web search domain restrictions to Claude Managed Agents sessions.

Significance 2: Meaningful trust-and-safety API plumbing for platforms/resellers, but incremental rather than verdict- or landscape-moving.

Sources

• primary Claude Platform release notes retrieved 2026-10-22

← Back to feed