In simple English
Your AI assistant remembers your code, but it forgets the people in your life. This tool gives your assistant a private memory box for names, roles, past agreements, and communication styles. All data stays in a single file on your computer. Nothing goes to the cloud or gets shared with companies. You keep full control over what is saved and what gets deleted.
What it is
A local-first Model Context Protocol (MCP) server and command-line tool built on SQLite. It lets AI agents query and update information about people through strict tools while keeping raw files private.
What it is good for
Solo builders who want their coding assistant or chat client to remember colleagues, clients, and partners without giving up personal data to a third-party service.
- Stores everything locally in a single SQLite file with no cloud dependency.
- Requires human approval before importing or saving new facts about anyone.
- Integrates smoothly with Claude Code, Claude Desktop, Cursor, and other MCP clients.
- Requires Python 3.11 and the uv package manager to install and run.
- Managing relationship graphs and imports adds manual upkeep overhead.
- Plaintext SQLite storage needs full-disk encryption to remain secure at rest.
manual contact notes and scattered chat history
Claude Desktop or Cursor
- mem0ai/mem068k stars
An AI memory layer that supports vector storage, but typically relies on cloud services or hosted infrastructure rather than a strict local-first setup.
- deepset-ai/haystack27k stars
A Python framework for joining language models, document search and tools into one working pipeline.
- Giskard-AI/giskard-oss5.9k stars
A Python library that writes tests for language-model programs and tries to break them on purpose.
- ggml-org/llama.cpp132k stars
A C and C++ library and command-line tool for running language models locally.
uv tool install people-context pctx init