Changelog

New Features:

Amazon Redshift Tools: Added toolkit for exploring Amazon Redshift databases and running queries. See the docs.

Revamp Human-in-the-Loop DX: We are introducing the RunRequirement class to handle HITL requirements. Now during HITL flows, you will find these objects in the Agent response or RunPaused event (if streaming). You can handle and resolve the HITL requirements via these objects. You can read more and see examples in the docs.
Yield run responses from continue_run: You can now pass yield_run_response to the continue run methods when streaming, to yield a RunOutput object at the end.
Async Databases with AgentOS Evals: You can now run Agno evals via the AgentOS with Agents and Teams that use an asynchronous database class.
Knowledge Retriever Update: Now passing run_context into get_relevant_documents_from_knowledge so custom retrievers has access to dependencies

Table creation with AsyncPostgresDb: fix a bug related to automatic table creation when using the AsyncPostgresDb class.
Memory Topics with SQLite: fix a bug getting all existing memory topics when using SQLite as database.

fix: propagate create_table_if_not_found in async Postgres _get_table by @xpander-moriel in #5600
fix: handle async dbs in OS eval endpoints by @manuhortet in #5602
feat: HITL update by @manuhortet in #5459
fix: Pass dependencies to custom knowledge retrievers by @willemcdejongh in #5601
fix: serve correct traces cookbook file by @RockPaperSmittie in #5603
feat: redshift connector toolkit [SDK-277] by @harshsinha03 in #5467
chore: add tests for streaming and db persistence with hooks by @manuhortet in #5604
chore: v2.3.7 by @manuhortet in #5606

Full Changelog: v2.3.6...v2.3.7