Skip to content
Lyuben Kikov edited this page Apr 14, 2026 · 39 revisions

Hey, welcome to ✨ Aperio wiki 🧐

Aperio gives your AI agents a shared memory — so they actually remember things across sessions, tools, and teammates.

🚧 Still building this out. Some pages are rough — bear with us.


Where do you want to start?

Never used Aperio before? Just looking for something useful? Building on top?
Start with Architecture to get the big picture of how it all fits together, then follow the Aperio-lite install guide to get up and running in minutes Jump straight to Commands for a full reference, or head to the MCP Tools Guide for hands-on examples you can follow along with Check out Custom Tools to extend Aperio with your own MCP tools, or UI Plugins if you want to tweak the interface itself

What's in here

🏗️ Architecture 🔌 MCP Tools 📚 MCP Tools Guide
How Aperio is structured under the hood — storage layers, memory flow, and the design choices that shaped it The full list of MCP tools Aperio exposes. Drop it into Cursor, Windsurf, or any MCP-compatible agent and start querying memory immediately Hands-on examples of Aperio in real workflows — what to call, when, and why. Good first stop if docs feel too abstract
🤖 AI Agents Comparison ⌨️ Commands 📦 Aperio-lite Install
Side-by-side breakdown of supported agents — capabilities, limits, and which one actually fits your use case Every CLI command Aperio ships with, what it does, and when to reach for it Step-by-step guide to getting the lightweight version running. Fewer dependencies, same core memory features
🧠 Memory Types 🔢 Embeddings 🛠️ Custom Tools
Episodic, semantic, and working memory — what each one stores, how they differ, and when Aperio uses which How Aperio vectorizes context and retrieves the right memories at the right time. Covers models, chunking, and similarity search Add your own MCP tools on top of Aperio. Covers structure, registration, and how to make them memory-aware
🧩 UI Plugins 👥 Team Setup
Extend or restyle the chat interface with custom plugin components Configure Aperio as a shared memory layer for your whole team — roles, access, and keeping context in sync across people and agents

Something broken or missing? Open an issue or check Troubleshooting.


Clone this wiki locally