Persistent Memory for
AI Agents

A backend service that solves the memory problem for AI agents. Store conversation history, replay sessions, fork conversations, and perform semantic search across your previous conversations.

Everything AI Agents Need

Built from the ground up to handle the unique requirements of AI agent memory management.

Persistent Storage

Store all messages with full context and metadata. Never lose conversation history again. Works with PostgreSQL, SQLite, and MongoDB. Choose the database that fits your stack.

Resumable Response Streams

Long lived response streams can survive client reconnects or user device switching.

Conversation Forking

Fork conversations at any message to explore alternative paths without losing the original.

Conversation Replay

Replay any conversation in the exact order messages occurred for debugging or analysis.

Access Control

User-based ownership and sharing with fine-grained permissions for secure multi-tenant and multi-agent use.

Encryption at Rest

Conversation entries are encrypted before storage so sensitive data stays protected even if the underlying store is compromised.

Full-Text & Vector Search

Search conversations and episodic memories by keyword or semantic similarity using integrated full-text and vector stores.

File Attachment Storage

Attach files, images, and documents to conversation entries with secure, token-based download links.

AI Stack Native

Designed for LangChain, LangGraph, Vercel AI SDK, Quarkus LangChain4j and Spring AI.