LeandroPG19/cuba-memorys

🧠 Persistent memory MCP for AI agents — Knowledge graph + Hebbian learning + Anti-hallucination. 12 tools, 1 dependency, zero manual setup.

39
/ 100
Emerging

Implements a PostgreSQL-backed knowledge graph with exponential decay memory (30-day halflife) and Hebbian learning via Oja's rule, storing entities, typed relations, and observations with automatic deduplication and contradiction detection. Provides hybrid search combining full-text (pg_trgm), semantic (pgvector HNSW), and graph traversal via entropy-routed reciprocal rank fusion, plus graph analytics (Leiden communities, Brandes centrality, PageRank) for anti-hallucination grounding. Built as an MCP server with 13 tools for knowledge engineering, error tracking, session awareness, and autonomous REM-sleep consolidation that runs on PostgreSQL 18+ with optional ONNX embeddings.

No Package No Dependents
Maintenance 13 / 25
Adoption 4 / 25
Maturity 9 / 25
Community 13 / 25

How are scores calculated?

Stars

6

Forks

2

Language

Rust

License

Last pushed

Mar 12, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/mcp/LeandroPG19/cuba-memorys"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.