synaptic-memory
Active·★ 29·MIT·Updated 2026-05-25
★ Trending★ RAG / Knowledge Base★ Memory & Context
Brain-inspired knowledge graph: spreading activation, Hebbian learning, memory consolidation.
Synaptic Memory is a knowledge graph and MCP tool server for LLM agents. It automatically ingests data from various formats into a structured graph, then provides 29 atomic tools for agents to explore. No LLM cost at indexing time.
#ai-agent#embedding#graph-database#hebbian-learning#knowledge-graph#llm#mcp#mcp-server
01
Features
01Auto-detects file format and generates domain profile
02Ingests text and structured data into knowledge graph
0329 atomic MCP tools for LLM agents
04Zero LLM cost at indexing
05Hybrid retrieval with BM25 + vector + graph reranking
02
Compatibility
SQLite
Native FTS5 + HNSW
Verified via docs
PostgreSQL
pgvector
Verified via docs
Kuzu
Embedded Cypher
Verified via docs
Memory
In-memory testing
Verified via docs
03
Quick start
1
$ pip install synaptic-memory
04
Use cases
↳Searching and reasoning over unstructured text corpora
↳Cross-document question answering
↳Absence proofs and paraphrased queries
05
Alternatives
Brave Search MCP★ 86.5k
Allow your AI Agent to search the real-time internet using Brave Search API. Essential for getting up-to-date information.
FunASR★ 16.6k
Industrial-grade speech recognition toolkit: 170x realtime, 50+ languages, speaker diarization, emotion detection, streaming, and OpenAI-compatible API.
Related searches
Comments
Log in to leave a comment
- RRiver BrownMay 25, 2026
Hebbian learning means the knowledge graph strengthens frequently co-accessed associations.
- RRobin LewisMay 6, 2026
More sophisticated than flat vector stores for complex knowledge structures.
- HHayden KimApr 17, 2026
Brain-inspired knowledge graph with spreading activation is a compelling architecture for agent memory.
- AAspen HarrisApr 1, 2026
Good for long-running agents where memory organization improves with use.