Security

How Recall keeps memory isolated per workspace

How Recall keeps memory isolated per workspace

How Recall keeps memory isolated per workspace

A walk through our isolation model, from encryption keys to retrieval-time permission checks.

AR

Marcus Lindholm

Jan 22, 2026 · 9 min read

·

Share

Shared memory is only useful if people trust it. That means a memory written in one workspace can never be read from another, and a person can never retrieve something they could not open in the original tool.

Isolation by default

Every workspace has its own encryption keys and its own index. There is no global index to query across customers, so there is nothing to leak through a ranking bug.

Permissions at retrieval time

We do not copy permissions at capture time and hope they stay current. Each retrieval checks the requester against the source system, so removing someone from a channel removes the memories from that channel immediately.

Audit and deletion

Every read and write is logged with the agent, the person and the reason. Deleting a source deletes its memories, and the deletion is recorded so you can show it happened.

AR

Written by Marcus Lindholm

Staff Engineer

Follow on X

LinkedIn

Give your AI a memory.

Deploy in minutes, keep your stack, and stop repeating yourself. Free for the first 10,000 memories.

SOC 2 Type II · GDPR ready · Self-host available

Give your AI a memory.

Deploy in minutes, keep your stack, and stop repeating yourself. Free for the first 10,000 memories.

SOC 2 Type II · GDPR ready · Self-host available

RECALL

The cross-tool memory layer for AI. Built for teams who ship with agents, copilots and automation.

© 2026 Recall Labs, Inc.

Created by Svorla

RECALL

The cross-tool memory layer for AI. Built for teams who ship with agents, copilots and automation.

© 2026 Recall Labs, Inc.

Created by Svorla

Get Template

Create a free website with Framer, the website builder loved by startups, designers and agencies.