Memento

What is Memento?

Memento is a local-first, LLM-agnostic memory layer for AI assistants. It runs an MCP server over a local SQLite file, enabling any MCP-capable AI assistant — Claude Desktop, Claude Code, Cursor, GitHub Copilot, Cline, OpenCode, Aider, and others — to read and write durable, structured memory about you, your work, and your decisions. It is fully offline by default with no outbound network calls and no vendor lock-in.

How to use Memento?

Install and initialize with npx @psraghuveer/memento init, which creates the database, runs migrations, and offers interactive setup. Then paste the printed MCP snippet into your AI client’s configuration and restart the client. After that, ensure the persona snippet reaches the assistant (auto-installed for many file-based clients; manual pasting for UI‑only clients). Verify setup with npx @psraghuveer/memento verify-setup. You can also run the server directly with npx @psraghuveer/memento serve.

Key features of Memento

• Five typed memory kinds: fact, preference, decision, todo, snippet
• Vector retrieval with FTS (first‑time model download ~110 MB)
• Bundled skills and persona snippets for assistant instruction
• Curated memory packs for seeding conventions and guides
• Local‑only dashboard web UI (npx @psraghuveer/memento dashboard)

Use cases of Memento

• Remember user preferences (e.g., preferred package manager) across fresh sessions
• Recall past decisions without re‑explaining rationale
• Seed a new project with conventions via starter packs (engineering‑simplicity, pragmatic‑programmer, etc.)
• Share curated memory packs with a team or across machines

FAQ from Memento

What are the runtime requirements for Memento?

Node.js ≥ 22.11 and a C/C++ toolchain (Xcode command‑line tools on macOS, build-essential on Debian/Ubuntu) for compiling better-sqlite3.

Where does Memento store its database?

By default in $XDG_DATA_HOME/memento/memento.db (typically ~/.local/share/memento/memento.db on POSIX). Override with --db or the MEMENTO_DB environment variable.

Does Memento send data over the network?

No outbound network calls by default. The only optional download is the vector‑embedding model (~110 MB) cached locally after the first search.

Which AI assistants work with Memento?

Any MCP‑capable client, including Claude Desktop, Claude Code, Cursor, GitHub Copilot, Cline, OpenCode, Aider, VS Code Agent mode, and others.

How does Memento handle memory conflicts?

Conflict management is documented in docs/guides/conflicts.md. Memento provides workflows for triaging and resolving conflicting memories.