Changelog

New Feature:

• Context Compression: Introducing compression of tool call results in a running context. Context Compression allows you to manage your agent context while it is running, helping the agent stay within its context window and avoid rate limits or decreases in response quality. This is still a beta feature. See the docs
• Memory Optimisation: Added the ability to summarize a user’s memories. This adds optimize_memories and aoptimize_memories to the MemoryManager as a way to summarize memories outside of agent runs. Only one strategy is currently supported, but this will be extended in future. Thie is still a beta feature. See the docs.
• Gemini File Search Support: Added support for Gemini File Search.
  • File Search Store Management: Create, list, get, and delete document stores
  • File Upload & Import: Direct upload with custom chunking configuration and metadata
  • Document Management: List, get, delete, and update documents with metadata filtering
  • Citation Extraction: Helper methods to extract and format source citations from responses
  • Async Support: Full async/await support for production workloads
  • Docs coming soon! In the mean time check out these cookbooks:
    • cookbook/models/google/gemini/file_search_basic.py
    • cookbook/models/google/gemini/file_search_advanced.py
    • cookbook/models/google/gemini/file_search_rag_pipeline.py

Improvements:

• output_schema on run/arun: override output_schema at runtime via the run() and arun() methods for both Agent and Team, as well as AgentOS API endpoints.
• API Key for AWS Bedrock: Added api_key support for AWS Bedrock. See the documentation.
• AWS Bedrock Claude: It is now more compatible with native Claude, which adds support for thinking models and caching.
• VertexAI Claude: It is now more compatible with native Claude, which adds support for thinking models and caching.
• Gemini 3 Thinking Level: Added support for thinking_level on Gemini

Bug Fixes:

• Async Knowledge Retriever: Fixed a bug where async knowledge retrieving would break due to missing async versions of get_user_message functions on Agent and Team
• Team Member Events: Ensure that member events are streamed if stream_member_events=True but stream_events=False on Team.
• MigrationManager Improvements: Fixed an issue preventing migrations from being applied for versions older than 2.3.0. Also adds support for a force migration flag
• StopAgentRun: Fixed the StopAgentRun exception from not correctly ending the agent run.
• PDF Image Reader: Fixed PDF Image Reader generating ID’s based on filename instead of using UUID.
• MCP tool_name_prefix: Made the tool_name_prefix parameter None by default to prevent it from adding _ by default as a prefix.
• AGUI:
  • Duplicate value for tool_call_id fixed
  • Missing assistant messages for tool calls fixed

What's Changed

• chore: Fix broken docs links by @willemcdejongh in #5501
• [docs] Fix broken links in examples and quickstart docs by @Bessouat40 in #5496
• fix: Async knowledge retriever by @willemcdejongh in #5503
• [fix] Correct inverted JSON output prompt logic in Team class by @aviscode in #5512
• feat: guardrails agentos demo by @anuragts in #5506
• fix: update how team events are handled for members by @dirkbrnd in #5509
• fix: how migrations are created by @dirkbrnd in #5522
• fix: serialize workflow step outputs by @ysolanky in #5526
• feat: output schema to run and arun [SDK-183] by @harshsinha03 in #5347
• Actually implement stopagentrun and retryagentrun sdk 260 by @uzaxirr in #5528
• fix: PDF Image reader generate UUID by @willemcdejongh in #5533
• feat: support for pymongo async sdk by @uzaxirr in #5465
• feat: implement memory optimization by @uzaxirr in #5229
• feat: tool call compression by @Mustafa-Esoofally in #5302
• fix: Reuse metadata for sql tables by @dirkbrnd in #5507
• feat: add Bedrock API key auth to AWS Claude by @naaa760 in #5497
• fix: claude aws by @ysolanky in #5513
• fix: Multiple AG-UI fixes by @Mustafa-Esoofally in #5407
• refactor: Remove unneeded code from AWS Claude model. This enables th… by @esw-amzn in #5489
• [feat] / enhancement: Leverage isinstance for child classes of MCPTools and MultiMCPTools by @Sebastiao-Assuncao in #5458
• feat: add thinking level to gemini by @ysolanky in #5544
• feat: add google gemini file search support. by @uzaxirr in #5381
• fix: Update flaky knowledge test by @willemcdejongh in #5548
• chore: release 2.3.3 by @ysolanky in #5524

New Contributors

• @Bessouat40 made their first contribution in #5496
• @naaa760 made their first contribution in #5497
• @esw-amzn made their first contribution in #5489

Full Changelog: v2.3.2...v2.3.3