Basic Memory is a knowledge management tool that connects AI assistants to persistent, searchable notes so conversations maintain context across sessions.
The tool solves the problem of AI assistants losing context between conversations by storing knowledge as Markdown files that both humans and AI can read and write. It works through a two-way sync mechanism where observations and wikilinks compound into a connected knowledge graph. Users can search notes semantically by meaning rather than keywords, and the system integrates with major AI clients and IDEs through the Model Context Protocol, allowing agents to discover and use tools progressively based on behavior hints rather than wasting context on trial-and-error.
Basic Memory suits developers and knowledge workers who want persistent context across multiple AI interactions without vendor lock-in. The local-first approach stores plain text on disk indefinitely, making it suitable for air-gapped environments or those prioritizing data ownership. A cloud option with cross-device sync is available for those who prefer managed hosting, while the open-source local installation remains free. The tool distinguishes itself through its two-way sync between human and AI edits, semantic search with optional reranking, and progressive tool discovery that reduces wasted context.
The project receives issues from both core maintainers and outside users, indicating adoption beyond the immediate team without overwhelming external support demands. Maintainers typically respond to new issues and pull requests within a day. Work in the issue tracker centers on bug fixes, enhancements, and cloud-related features.
