Every AI conversation starts from zero. Second Brain gives Claude, ChatGPT, Cursor, and any MCP client persistent memory — self-hosted on Cloudflare's free tier.
Every new conversation, you re-paste context. Your tech stack, your decisions, what you agreed on last Thursday. The AI is brilliant in the moment and completely blank at the start of the next session.
The built-in memory features that exist are black boxes. You can't search them, audit them, or control what gets stored. They occasionally remember something useful.
Second Brain is a memory layer you actually control.
Click the deploy button. Cloudflare provisions your D1 database, Vectorize index, and Worker automatically. No configuration needed.
Add your Worker URL to Claude Desktop, Claude Code, or claude.ai. Works with Cursor and any MCP-compatible client. One URL, every tool.
Claude stores context automatically during conversations. At the start of every new session, it recalls what's relevant. You stop re-explaining.
Finds memories by meaning, not keywords. "Users drop off at checkout" surfaces when you search "conversion problems."
Browse, search, and manage your memories from a clean UI at your Worker URL. No API knowledge required.
Everything lives in your own Cloudflare account. No third party ever touches your memories.
Claude Desktop, Claude Code, claude.ai, Cursor, and any MCP-compatible client. One memory layer, every tool.
Long notes split into overlapping segments. Each section gets its own vector so specific parts surface precisely.
Near-exact matches blocked before storing. Your memory stays clean without accumulating repeated context.
No credit card required for personal use.
Free, open source, and self-hosted. Your memories stay in your own Cloudflare account.