Memex: your context, owned by you

Memex: your context, owned by you is an open-source Model Context Protocol server that stores user context in a Supabase database for AI agents.

Memex is an open-source context layer that you own, keeping durable information about you and your work shared across every AI tool. It runs as an MCP server on your own Supabase project, letting any agent use and update your preferences, decisions, corrections, and background information. You control your own database and keys, paying only pennies a month for embeddings and optional processing. It features hybrid search, scheduled sync tasks, and an optional nightly dreaming job that organizes and connects your memories while keeping your data safe and secret-free. It works with agents that support MCP over HTTP, including Claude, Muse, Cursor, and VS Code.

Memex: your context, owned by you at a glance

Best for: Software engineer
Runs on: Web, VS Code, self-hosted, open source
Pricing: Free and open source; database and embeddings costs are pennies a month on Supabase
Released: November 2026
Category: AI agent tool
Website: predsymod.github.io

Key features

Stores durable personal and work context in your own database
Runs as an MCP server on your own Supabase project
Blends meaning and exact words in search results
Syncs context across multiple AI agents
Runs nightly jobs to organize and connect memories
Flags and hides secrets automatically
Retracts and restores saved notes easily

Use it for

Share context across multiple AI coding agents
Store work preferences and decisions in your own database
Sync updates and learned facts automatically using scheduled tasks
Organize and connect memories with nightly dreaming jobs

Alternatives to Memex: your context, owned by you

ContextMCP — Maintains fresh, accurate context for AI agents using AST-aware chunking.
memU — It gives every connected agent a shared and inspectable memory layer.
claude-mem — It shares persistent, searchable agent memory across sessions and devices.
MemPalace — Achieves high recall with local-first storage and zero API calls.
Memmy — Shares one persistent memory foundation across your different AI tools.
MemMachine — Provides long-term persistent memory for LLM-based agents.
Context7 — Provides up-to-date documentation for AI agents and code editors.
Supermemory — Integrates long-term memory into AI coding agents with a single prompt.
OpenViking — Unifies agent memory, knowledge RAG, and skills into a single directory structure.
Agentmemoryrepo — Enables agents to navigate memories like a wiki and share knowledge across sessions.
ProjectMem — Records project history locally to prevent AI from repeating failed work.
Mem0 — It provides drop-in persistent memory for AI agents without changing pipelines.

All Memex: your context, owned by you alternatives