Claude API adds user profiles, so a safety action can hit one end customer instead of your whole org

Loading the board

Anthropic's October 20 release notes add /v1/user_profiles endpoints for platforms, gateways and resellers, plus organization webhooks (beta) that fire when Anthropic restricts or bans one of those profiles.

If you run a platform or gateway on the Claude API, or resell access, you can now tell Anthropic which of your end customers each request is for. Anthropic says a safety enforcement action can then apply to that one customer instead of your whole organization. The SDK and the ant CLI support the new /v1/user_profiles endpoints.

The same release adds organization webhooks to the Admin API, in beta behind the webhooks-2026-10-07 header. They cover two events: Anthropic restricting or banning one of your user profiles, and that restriction ending. Access goes through your Anthropic account team.

Inference hooks, which let a Claude Enterprise org's own security server allow or deny each prompt, also left beta on October 20.

Why it matters for builders

For anyone reselling Claude, one abusive customer could get the whole account restricted. With profiles, the action can land on that customer, and the webhook tells you about it so you can act on your side.

Try this

If you resell Claude API access, create one user profile per end customer with the new /v1/user_profiles endpoints (the SDK and the ant CLI both support them), so an enforcement action can apply to that customer instead of your whole organization.

Claude API adds user profiles, so a safety action can hit one end customer instead of your whole org

Why it matters for builders

Try this

Sources

More AI news

Claude API adds user profiles, so a safety action can hit one end customer instead of your whole org

Why it matters for builders

Try this

Sources

Get stories like this every Monday

More AI news