Overview

zero-memory is a shared, persistent memory for LLM coding agents. It is exposed through the Model Context Protocol (MCP), so any MCP-compatible tool — Claude Code, Cursor, Codex CLI, VS Code, and others — reads and writes the same memory instead of rebuilding context from scratch in every session.

Why

Every new chat session starts from zero. Most existing memory layers are single-user or hide team sharing behind a hosted cloud. zero-memory targets the open niche: self-hosted, open-source, multi-user team memory with access control enforced in the database.

Key ideas

• Two-tier memory — private fragments visible only to their author, plus selectively shared fragments with provenance (who, when, from which conversation). Sharing is always an explicit action.
• Explicit scope membership — private personal memory plus owner-namespaced, shareable project and team scopes. A matching project name never grants access, the session's project is the write default, and a write with no resolvable target is refused rather than mis-filed. See Scopes and isolation.
• Several write paths — deliberate agent writes, terminal quick capture, deterministic import, repository bootstrap, and opt-in transcript ingest. Transcript extraction is disabled by default.
• MCP-first — memory, briefing, hygiene, rules, import/export, and metrics tools work in any MCP-compatible client. See the memory model.
• Network-served standing rules — promoted rules use native MCP instructions where the client can carry them and build_context everywhere; hooks add automatic delivery timing and offline resilience.
• Postgres (Supabase) backend — hybrid pgvector + full-text retrieval, an entity graph as edge tables, Row Level Security for scoping, and Realtime for live team updates.

How the docs are organized

• Getting Started — connect a client to a running instance, or deploy your own stack.
• Concepts — the memory model and the design decisions behind it.
• Guides — day-to-day workflows: briefings, import/export, session receipts.
• Dashboard — the web UI for browsing and curating memory.
• Memory Hygiene — how the store stays clean over time.
• Platform — architecture, observability, testing, and internals.