Skip to content

Supermemory SOPLong-term memory infrastructure for AI agents

Operating procedures, integration recipes, and complete API reference โ€” mirrored from the official Supermemory documentation.

What's inside โ€‹

This is a comprehensive standard-operating-procedures companion to the Supermemory platform. Every section is mirrored locally so it can be searched, browsed, and queried even offline โ€” and the AI assistant (bottom-right corner) is wired into a vector store of all 121 documentation pages.

Quick Navigation โ€‹

Where to start

Coverage at a glance โ€‹

SectionPagesWhat's covered
Getting Started7Intro, quickstart, overview, authentication, vibe coding, changelog, add-memories
Concepts8How it works, memory vs RAG, graph memory, super RAG, content types, filtering, customization, user profiles
Memory API5Memory operations, document operations, search, user profiles
MemoryBench11The open benchmark โ€” installation, CLI, architecture, providers, integrations, MemScore, GitHub
Connectors9Google Drive, Notion, OneDrive, Slack, manual mode, overview
Integrations22AI SDK, LangChain, Mastra, OpenAI, Vercel AI, CrewAI, Pipecat, Letta, and more
SMFS10Supermemory File System โ€” pricing, security, supported models, write to memory
MCP3Personal MCP, MCP clients, hosted MCP
Migration3From Mem0, Letta, Zep
Cookbook7Real-world recipes โ€” Discord, FastAPI, Lovable, more
API Reference36Documents, connections, memories, search, management

Built as an internal reference companion.