Shared memory across tools
Every connected client reads the same vault, so context captured in one AI tool is recallable in the next.
Give your AI agents a shared, searchable long-term memory over MCP.
1AIVault ships its own MCP server, so any connected AI client can reach the vault through standard tool calls. Agents can run keyword and semantic searches, pull recent entries, fetch a specific memory by id, follow related links, and save new memory back — the same recall surface you use in the app, exposed to Claude, Cursor, Codex, and the other supported clients over a single local connection.
This turns a stateless assistant into one with persistent, cross-tool memory. A conversation you had in one client becomes recallable context in another, because they all read from the same local vault. Nothing routes through a vendor cloud: the MCP server runs on your machine, and every call it serves is recorded in Memory Reads, so agent access to your memory stays both powerful and fully accountable.
Every connected client reads the same vault, so context captured in one AI tool is recallable in the next.
Search, recent, get-by-id, related, and save tools give assistants the same retrieval power you have in the app.
The MCP server runs on your device and logs each call in Memory Reads, so agent access stays private and auditable.
Install the 1AIVault MCP server into a supported client from Connection → Install MCP.
Restart the client and ask it to search your vault.
The agent calls recall tools like vault_search, vault_brain_query, or vault_recent to read memory.
Confirm the call landed in Connection → Memory Reads.
Through the built-in MCP server. Once installed into a client, the agent can call recall tools to search, fetch, relate, and save vault memory over a local connection.
Supported MCP clients include Claude Desktop, Claude Code, Cursor, Cline, Codex, Gemini, Antigravity, ChatGPT, and OpenCode, all reading from the same local vault.
Dashboard view shows a vault summary plus a live activity feed of every read, write, and classification across all connected AI tools. Deep links via `aivault://` open the app from anywhere.
Learn moreEvery entry detail header shows 'Last used X ago by Source · Nx this week' so you know which AI tools are touching a memory and how often.
Learn moreLocal embeddings — via Xenova Transformers (ONNX), llama.cpp (GGUF), or a remote embedding API — so AI clients can recall memories by meaning, not just keywords.
Learn moreStart free, import real conversations, and reuse your memory across every AI agent you already use.