brainctl

Overview

brainctl is a Python-based cognitive memory system for AI agents, providing episodic, semantic, and procedural memory with MCP server integration, full-text search, and optional vector search capabilities.

Capabilities

• • database integration
• • mcp client integration

Best For

MCP-compatible cognitive memory tool for AI agents, built in Python to manage and search agent memory.

Decision Snapshot

• • GitHub stars: 21
• • Forks: 3
• • Source provenance count: 1
• • Active signal: Recently published.
• • Last seen: 4/5/2026
• • Published: 4/3/2026

Installation / Setup

pip install brainctl              # core && pip install brainctl[mcp]         # with MCP server && pip install brainctl[vec]         # with vector search (sqlite-vec) && pip install brainctl[all]         # everything

Usage

{ && "mcpServers": { && "brainctl": { && "command": "brainctl-mcp"

Features

• • **Query the brain, don't inject it.** Don't dump memory into every system prompt. Search when relevant.
• • **Use oneline for routine lookups.** Full JSON is for debugging. Agents need facts, not metadata.
• • **Set --budget on automated queries.** Cron jobs and heartbeats should cap their own output.
• • **Limit scope.** --tables memories skips events/context. --category convention narrows further.
• • **Let salience filtering work.** --min-salience 0.1 drops noise that wastes tokens downstream.
• • database integration
• • mcp client integration

Use Cases

• • Supports capabilities such as: database integration; mcp client integration.
• • Common usage themes: ai, agents, memory, mcp.

Supported Clients / Integrations

• • database integration
• • mcp client integration

Prompt Examples

Notes / Requirements

• • Primary language: Python
• • License: MIT
• • Parser coverage score: 1.00
• • Source feeds: PyPI RSS + JSON API
• • Topic cluster: data-infra

Official Links

• Website
• Repository

Source Information

Related MCP Tools