- Generalized OpenAPI spec reference in client.py - Removed personal IP addresses and user IDs from README - Added MEM0_PREFETCH_LIMIT and MEM0_PREFETCH_SCORE_THRESHOLD to documentation - Updated example configurations with generic values
- Self-hosted Mem0 integration (no cloud dependency) - Async prefetch with ~40ms latency - Automatic context injection via pre_llm_call hook - Circuit breaker for server resilience - Full tool support: profile, search, conclude