0.148.0
📦 crewaiView on GitHub →
✨ 5 features🐛 5 fixes🔧 8 symbols
Summary
This release introduces comprehensive Agent evaluation tools and neatlogs integration, alongside critical fixes for agent knowledge handling and Task parameter configurations.
✨ New Features
- Introduced Agent evaluation functionality including experiment and regression testing methods
- Implemented thread-safe AgentEvaluator with event emission support
- Enabled evaluation support for single Agent and LiteAgent instances
- Added integration with neatlogs for enhanced logging
- Added crew context tracking for LLM guardrail events
🐛 Bug Fixes
- Fixed agent knowledge handling logic
- Resolved missing create_directory parameter in Task class
- Updated LLM comparison logic to use BaseLLM class instead of LLM
- Corrected WorkOS environment ID to use production values
- Added SQLite FTS5 support to test workflows
🔧 Affected Symbols
AgentEvaluatorAgentLiteAgentTaskBaseLLMLLMAgent.kickoffneatlogs