MenteDB

MenteDB is a purpose built database engine for AI agent memory. Not a wrapper around existing databases, but a ground up Rust storage engine that understands how AI/LLMs consume data.

Core features:

• Automatic Memory Extraction LLM powered pipeline extracts structured memories from raw conversations
• Entity-Centric Memory Extracts typed entities (person, pet, place, event, item, organization) with structured attributes. Entity resolution merges attributes across mentions. Graph edges link memories to the entities they reference
• Adaptive Multi-Pass Retrieval Engine-level 3-pass search (instant recall ? active search ? deep dig) with progressively increasing depth, reciprocal rank fusion, and entity graph expansion
• Write Time Intelligence Quality filter, deduplication, and contradiction detection at ingest
• LLM Powered Cognitive Inference CognitiveLlmService judges whether new memories invalidate, update, or are compatible with existing ones (supports Anthropic, OpenAI, Ollama)
• Bi-Temporal Validity Memories and edges carry valid_from/valid_until timestamps. Temporal invalidation instead of deletion. Point-in-time queries via recall_similar_at(embedding, k, timestamp)
• Attention Optimized Context Assembly Respects the U curve (critical data at start/end of context)
• Belief Propagation When facts change, downstream beliefs are flagged for re evaluation
• Delta Aware Serving Only sends what changed since last turn (90% reduction in memory retrieval tokens over 20 turns)
• Cognitive Memory Tiers Working, Episodic, Semantic, Procedural, Archival
• Knowledge Graph CSR/CSC graph with BFS/DFS traversal and contradiction detection
• Memory Spaces Multi agent isolation with per space ACLs
• MQL Mente Query Language with full boolean logic (AND, OR, NOT) and ordering (ASC/DESC)
• Type Safe IDs MemoryId, AgentId, SpaceId newtypes prevent accidental mixing
• Binary Embeddings Base64 encoded storage, 65% smaller than JSON arrays
• Local Candle Embeddings Zero config semantic search using all-MiniLM-L6-v2 (384 dims), no API key required
• gRPC + REST + MCP Three integration paths for any use case