Anthropic Python SDK v2.4.0 Adds Analytics, Spend Limits, and New Error Types in API

These changes increase the flexibility and robustness of applications built on the Anthropic API. Developers now have direct support for analytics queries, spend control, granular RBAC actions, and better metadata management for keys and workspaces. New error types help catch session-specific issues related to tool and compaction limits, improving user and system feedback. Nullable spend_limit support and dropped legacy fields provide clearer, less ambiguous responses. These are valuable improvements for teams monitoring API usage, managing quotas, or building analytics dashboards.

What can the developer do now?

Developers can update to v2.4.0 to access new analytics and spend limit APIs, handle the newly introduced error types in their session logic, and make use of finer filtering for analytics reporting. Projects that rely on API responses should now account for nullable spend_limit values. Integrating user_profile_id will streamline managing access keys and workspace resources. No breaking changes are noted, so upgrades are non-disruptive.