"Pocket" matching MCP servers:

Enables LLMs to read a user-owned, local-first memory store kept in a portable folder (USB, SSD or laptop), with per-profile permissions limiting what each client can see and access cut off when the pocket is locked or unplugged. Also lets models propose new memories for the user to review and approve, keeping provenance and keeping canonical memory free of direct model writes.

Enables an AI assistant to keep every conversation verbatim in a local SQLite store and recall the most relevant past turns on demand, using hybrid keyword and embedding search plus time-awareness. Results are packed into a small token budget (2k–8k tokens) so long histories fit modest context windows, with secrets masked before storage.

Persistent memory storage using SQLite for AI agents to store, search, update, and manage contextual memory via the Model Context Protocol.

Enables AI assistants to save, retrieve, and manage research content using ChromaDB vector storage with semantic search, topic organization, and automatic deduplication powered by OpenAI embeddings.

A privacy-first personal context engine that ingests personal photos and message exports, extracts cited facts, and serves a queryable profile over MCP for natural-language questions like 'when did I last see Sarah?'.