v0.4.0 Latest
New Features
- Added
memory_consolidatetool for merging old memories based on decay threshold - New
knowledge_graph_buildandknowledge_graph_querytools for relationship mapping - Introduced
codebase_generatetool for AI-powered code generation from context - Added batch operations:
memory_batch_store,memory_batch_tag,memory_batch_delete - New
agent_handofftool for seamless context transfer between agents - Added
memory_priority_scorefor intelligent memory ranking - New conversation management:
conversation_start,conversation_add_turn,conversation_save
Improvements
- Memory search now uses hybrid vector + keyword matching for 40% better relevance
- Reduced average response time from 12ms to 8ms across all operations
- Improved
llm_context_windowtoken estimation accuracy to 99%+ - Vector capacity increased from 25K to 50K vectors per instance
- Added automatic deduplication during memory storage