> ## Documentation Index
> Fetch the complete documentation index at: https://labs.prompthon.io/llms.txt
> Use this file to discover all available pages before exploring further.

# April 2026 Portable Assistant Memory Watch

<div className="not-prose my-4 rounded-md border border-gray-200 bg-gray-50 p-2 text-sm dark:border-gray-800 dark:bg-gray-900/40">
  <div className="mb-2 px-1 text-xs font-semibold uppercase tracking-wide text-gray-500 dark:text-gray-400">Need help?</div>

  <div className="flex flex-wrap gap-2">
    <a className="inline-flex items-center gap-2 rounded-md border border-gray-200 bg-white px-3 py-1.5 font-medium text-gray-700 no-underline shadow-sm hover:border-gray-300 hover:bg-gray-50 dark:border-gray-800 dark:bg-gray-950/60 dark:text-gray-200 dark:hover:bg-gray-900" href="https://discord.gg/sDE2HhGTg4" target="_blank" rel="noreferrer">
      <Icon icon="discord" iconType="brands" size={14} />

      <span>Ask in Discord</span>
    </a>

    <a className="inline-flex items-center gap-2 rounded-md border border-gray-200 bg-white px-3 py-1.5 font-medium text-gray-700 no-underline shadow-sm hover:border-gray-300 hover:bg-gray-50 dark:border-gray-800 dark:bg-gray-950/60 dark:text-gray-200 dark:hover:bg-gray-900" href="https://github.com/Prompthon-IO/agent-systems-handbook/issues/new/choose" target="_blank" rel="noreferrer">
      <Icon icon="github" iconType="brands" size={14} />

      <span>Open a GitHub issue</span>
    </a>

    <a className="inline-flex items-center gap-2 rounded-md border border-gray-200 bg-white px-3 py-1.5 font-medium text-gray-700 no-underline shadow-sm hover:border-gray-300 hover:bg-gray-50 dark:border-gray-800 dark:bg-gray-950/60 dark:text-gray-200 dark:hover:bg-gray-900" href="https://github.com/Prompthon-IO/agent-systems-handbook/blob/main/SUPPORT.md" target="_blank" rel="noreferrer">
      <Icon icon="life-ring" size={14} />

      <span>Support guide</span>
    </a>
  </div>
</div>

## Summary

Google's April 29, 2026 Gemini update pushes assistant memory into a more
portable shape: remembered preferences, imported memory summaries from another
assistant, and full chat-history transfer now sit behind the same
personal context story.

For handbook readers, the useful shift is not "better personalization" in the
abstract. It is that imported personal context is becoming a first-class input
surface that sits next to retrieval, working memory, and durable artifacts.

## Why It Matters

The lab already explains memory and retrieval as a systems boundary, but this
product signal sharpens a newer question: what exactly counts as memory when a
user can import a summary or a ZIP archive from another provider?

That matters because these inputs behave differently:

* retrieval pulls from external sources when needed
* working memory helps the current task stay coherent
* durable artifacts preserve explicit outputs such as notes or decisions
* imported personal context brings prior preferences, relationships, and
  conversational history into a new assistant

Treating all four as one vague "memory" bucket makes review, consent, and
deletion much harder to reason about.

## Evidence And Sources

* [Gemini launches new personalisation features in the UK](https://blog.google/company-news/inside-google/around-the-globe/google-europe/united-kingdom/gemini-launches-new-personalisation-features-in-the-uk/):
  Google says Gemini memories can learn from past conversations, that the
  memory setting is on by default, and that users can import both memory
  summaries and full chat-history exports from other AI apps.

## Signals To Watch

* Whether assistant products keep imported personal context separate from
  retrieval inputs and durable artifacts in their internal models and user
  controls.
* Whether default-on memory features lead to clearer review, deletion, and
  scope-management surfaces.
* Whether cross-provider migration becomes a standard product expectation
  rather than a one-off import tool.
* Whether enterprise products adopt the same portability pattern or keep it
  limited to consumer assistants.

## Editorial Take

This signal belongs in `radar/`, not in evergreen memory prose yet. The durable
takeaway is the boundary it exposes:

* imported preferences are not the same as retrieval
* imported chat history is not the same as a durable work artifact
* personalization settings are now part of the trust surface, not only a UX
  convenience

Future refreshes to memory-related handbook pages should keep those categories
separate instead of collapsing them into a generic memory stack.

## Update Log

* 2026-04-29: Added a radar note on imported personal context, memory
  portability, and chat-history transfer as a new assistant-design boundary.