Tenure: A local AI memory system that stores both facts and actionable instructions for LLMs
By
Jeff Flynt
Good intentions, undercooked execution. The bake is missing.
Summary
Tenure is a local AI memory system that goes beyond storing facts by also storing instructions on what to do with those facts. Unlike most LLM memory systems that only store factual information, Tenure includes a "why_it_matters" field that converts observations into actionable instructions for the model. It features precision retrieval over similarity search, is fully local and encrypted, and ensures all beliefs are visible, editable, and correctable. The system's retrieval claims are documented in a reproducible arXiv paper.
Key quotes
· 4 pulledYou already told the model you use Fastify. That you want clarifying questions before long responses. That you chose the MongoDB raw driver over Mongoose for a specific reason. Next session, it has no idea and meets you as a total stranger.
Most LLM memory systems store facts. Tenure stores what to do with them.
Every belief carries a why_it_matters field that converts observations into instructions the model acts on directly, no additional inference required.
Fully local, encrypted at rest, nothing leaves localhost. Every belief is visible, editable, and correctable, the model isn't building a hidden profile.
You might also wanna read
Ensue Memory Network: Persistent Knowledge System for AI Conversations
Ensue is a persistent memory network for AI that allows LLMs to maintain knowledge across conversations, creating a growing knowledge tree t
github.com·5mo ago
Letta Code: A Memory-First Coding Agent for Persistent Learning Across Sessions
Letta Code is a memory-first coding agent designed for long-lived, persistent AI agents that learn and improve across multiple sessions rath
letta.com·5mo ago