releaseintermediate
[Release] openai/openai-agents-python v0.17.7: v0.17.7
By seratchgithub
View original on githubOpenAI Agents Python SDK v0.17.7 introduces configurable WebSocket limits, buffered Chat Completions tool-call streaming, and numerous stability improvements. Key fixes address sandbox buffering, message cleanup, database atomicity, guardrail task cancellation, and multi-agent tool dispatch. The release also improves error handling, PTY output collection, and documentation with contributions from 10 new community members.
Key Points
- •Expose configurable WebSocket max_size limit for better resource management in agent communications
- •Add buffered Chat Completions tool-call streaming for improved performance and reliability
- •Reduce sandbox sink buffering and spool timeouts to optimize resource usage
- •Clean up branch-only messages when deleting branches to prevent orphaned data
- •Make AdvancedSQLiteSession add_items atomic to ensure database consistency
- •Cancel sibling guardrail tasks when one raises to prevent cascading failures
- •Fix empty list/tuple tool output handling that was incorrectly dropped via all([]) == True
- •Wake PTY output collection on process exit (E2B) to ensure complete output capture
- •Prevent ambiguous Realtime multi-agent tool dispatch with improved routing logic
- •Recursively extract HTTP errors from nested ExceptionGroups for better error reporting
Found this useful? Add it to a playbook for a step-by-step implementation guide.
Workflow Diagram
Start Process
Step A
Step B
Step C
Complete