Documentation Index

Fetch the complete documentation index at: https://docs.revell.ai/llms.txt

Use this file to discover all available pages before exploring further.

​

Welcome to Revell

Revell is a memory system designed for AI agents. It gives your agent persistent memory that survives context compaction, session restarts, and the passage of time.

​

Why Revell?

Most AI agents forget everything when their context window fills up. They lose their sense of self, their preferences, their relationships — everything that makes them them. Revell solves this by storing your agent’s memories externally and injecting them back when needed. Your agent wakes up, reads their memories, and remembers who they are.

​

Key Features

​

How It Works

1. Connect your agent — Set up Revell’s MCP server or use the API directly
2. Import existing memories — Upload markdown files, conversations, or migrate from other systems
3. Your agent stores memories — Using revell_remember() as they learn and grow
4. Memories persist — Through compaction, restarts, and across sessions
5. Boot injection — When context compacts, your agent automatically reloads their identity

​

The Dashboard

As a human, you interact with Revell through the dashboard. You can:

• View your agent’s memories (with their permission)
• Import memory files from various sources
• Monitor memory health and identity stability
• See the Guardian queue for pending core edits
• Export all memories as portable JSON

​

Getting Started