Zenbrain MCP server
{
"mcpServers": {
"zensation-zenbrain": {
"command": "npx",
"args": [
"-y",
"@zensation/mcp@latest"
]
}
}
}{
"mcpServers": {
"zensation-zenbrain": {
"command": "npx",
"args": [
"-y",
"@zensation/mcp@latest"
]
}
}
}{
"mcpServers": {
"zensation-zenbrain": {
"command": "npx",
"args": [
"-y",
"@zensation/mcp@latest"
]
}
}
}{
"servers": {
"zensation-zenbrain": {
"command": "npx",
"args": [
"-y",
"@zensation/mcp@latest"
]
}
}
}{
"mcpServers": {
"zensation-zenbrain": {
"command": "npx",
"args": [
"-y",
"@zensation/mcp@latest"
]
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @zensation/mcpversion 0.1.3
What it does
ZenBrain is a TypeScript memory system for AI agents that implements seven neuroscience-inspired memory layers using algorithms like spaced repetition (FSRS), Hebbian learning, and sleep consolidation. It stores memories locally in SQLite with no account required, and is designed for agents that need to retain and consolidate information across conversations.
- Build an AI agent that remembers facts, procedures, and emotional context from past interactions without losing information between sessions.
- Implement adaptive memory consolidation where frequently-used knowledge strengthens and rarely-used knowledge fades following human forgetting curves.
- Create a multi-turn conversational system where the agent's reasoning improves by drawing on consolidated episodic and semantic memories from prior exchanges.
Tools this server exposes
| Tool | Does |
|---|---|
| FSRS | Spaced repetition scheduling algorithm for optimizing memory recall intervals. |
| vmPFC-FSRS | Advanced variant of spaced repetition incorporating ventromedial prefrontal cortex principles. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.