Persistent Memory — LM Studio Plugin

Give your local LLM a long-term brain. This plugin stores facts, preferences, projects, and notes across conversations in a local SQLite database — zero cloud, zero external services, fully portable.

Retrieval Scoring (Inspired by SRLM)

Memories are ranked by a composite score blending four signals (inspired by the SRLM paper which showed that combining multiple uncertainty signals outperforms any single one):

Signal	Weight	What it measures
TF-IDF Similarity	55%	Semantic relevance to the current query
Recency Decay	20%	Exponential decay based on last access time
Confidence	15%	How certain we are about this fact (0–1)
Access Frequency	10%	How often this memory has been surfaced

Decay follows score = 2^(-days/halfLife) — memories that stop being accessed gradually fade, just like human memory.

Memory Scopes

Scope	Persistence	Use for
`global` (default)	Forever, all chats	User facts, preferences, standing instructions
`project`	Forever, project-filtered	Project-specific context, repo details, team info
`session`	Until LM Studio closes	Temporary context, scratch notes, current-task state

Global memories are injected into every conversation automatically. Project memories are persisted to SQLite but only surface when that project is referenced. Session memories live entirely in memory — they're never written to disk and vanish when the plugin reloads. This lets you store throwaway context without polluting your permanent knowledge base.

Memory Categories

Category	Use for
`fact`	Things about the user: name, job, skills
`preference`	Likes, dislikes, coding style, UI choices
`project`	Current work, repos, goals
`note`	Free-form memos
`instruction`	Standing instructions ("always use TypeScript")
`relationship`	People, teams, organizations
`context`	Situational context ("interviewing at X")

Project-Scoped Memory Guidance

When Active Project is set, durable repo knowledge should be written into that project's namespace instead of global memory.

Scope split

global: user-wide preferences and standing instructions
project: repo-specific knowledge, architecture decisions, workflow rules, team conventions, recurring issues
session: temporary task state and scratch context

Important limitation

Setting Active Project and enabling AI extraction is only sufficient if the write path also stores extracted memories with scope: "project" and the configured project slug. This plugin now applies that rule to auto-extracted chat memories.

Project ingest / bootstrap

Use the bootstrap flow to store distilled project memories, not raw file dumps. Good bootstrap entries include:

architecture invariants
important commands
service boundaries
persistent constraints
known failure modes
active goals

Run a project ingest when:

a repo is connected for the first time
architecture changes materially
major tickets land
conventions change

Privacy

All data stays on your machine. The memory database is a single file at ~/.lmstudio/plugin-data/persistent-memory/memory.db. You can:

Back it up: Just copy the file
Move it: Change the storage path in settings
Delete it: Remove the file to start fresh
Inspect it: Open with any SQLite browser

License

MIT