Memory Vault AI Documentation

Welcome to the official documentation portal for Memory Vault AI. This site helps developers understand, integrate, deploy, and extend Memory Vault AI quickly.

Start in 60 Seconds

1. Pick your path from the quick links below.
2. Follow the SDK guide for embedded use, or deployment guide for service mode.
3. Use API and spec docs when implementing production integrations.

Quick Links

Goal	Best page
Save and recall memory in Python	SDK Guide
Configure runtime behavior	Configuration
Run with Docker or production topology	Deployment
Integrate with coding assistants	MCP Integration
Benchmark performance	Benchmarking
Extend memory classification	Plugin System Guide

What You Can Do With Memory Vault AI

• Understand architecture and memory flow end-to-end.
• Integrate Memory Vault through SDK, REST API, CLI, or MCP.
• Configure storage and runtime behavior for local or production use.
• Extend memory routing using custom memory type plugins.

Quick Start

import asyncio

from memory_vault import MemoryLayer


async def main() -> None:
    memory = MemoryLayer(user_id="alice")
    await memory.save("I prefer concise answers with examples.")
    result = await memory.recall("How should I respond to the user?")
    print(result.prompt_block)


asyncio.run(main())

Technical Reference

• REST API Spec
• SDK API Spec
• Architecture
• Memory Logic
• Database Schema
• Plugin System Spec

Architecture Decisions

• Review ADR entries in the left navigation to understand design tradeoffs.

Public Hosting via GitHub Pages

This docs site is designed for GitHub Pages deployment via GitHub Actions. After Pages is enabled in repository settings, docs are available at:

• https://zidanmubarak.github.io/Memory-Vault-AI/

Build This Website Locally

pip install -e ".[docs]"
python -m mkdocs serve

Static build output:

python -m mkdocs build