Your private memory graph. Sign in to open your brain.
// results appear here
Give your own scripts and agents access to this brain — no browser, no shared secrets. This key maps only to your tenant; the tenant is the key's identity.
……
Scheduled · activation pending
Your brain is scheduled for a nightly snapshot (~02:00) to per-tenant, encrypted object storage (Cloudflare R2). Backup credential activation is in progress — until it lands, snapshots are queued but not yet uploading. You'll see the last successful backup here once it's live.
Losing sleep over it? You can export your whole graph any time from the Connect API.
Two steps: connect brainbox to your AI, then paste one instruction so it recalls and records as you work — the value shows up in the conversation, not on this page.
Run this once, then sign in with your brainbox account when the browser opens:
claude mcp add --transport http brainbox https://brainbox-mcp.thexi.dev/mcpAdd a remote / custom MCP connector pointing at https://brainbox-mcp.thexi.dev/mcp and sign in with your brainbox account. Desktop builds that are stdio-only can bridge with the mcp-remote shim.
Paste this where your AI reads every session — Claude Code CLAUDE.md, a Claude / claude.ai Project's custom instructions, or Cursor Rules:
Before answering anything about me or my work, call `recall` with the key terms to load what you already know about me. When you learn something durable — a decision, preference, person, project, or fact worth keeping — call `remember` to save it as one short factual sentence. Skip small talk.
That's the whole loop: your AI reads your brain before it answers and writes back what matters. It gets sharper every session.
No browser login? Point the same MCP endpoint https://brainbox-mcp.thexi.dev/mcp at your per-tenant key from the Connect tab — set the header Authorization: Bearer <bbk_key> directly (OpenAI Codex, Cursor, CI all support this natively; no mcp-remote needed). Or skip MCP entirely: POST /cypher with the same key.