# While you were away — the digest

When the desktop window regains focus after you've been away (5 minutes or more), Twico shows a digest at the bottom of the [Home tab](/twico/the-interface/home.md). It's a one-paragraph summary of what happened — written as if a colleague were telling you what landed in your inbox — plus a per-conversation breakdown.

The digest is **inline, gray, and subtle** by design. It's information, not a popup. It doesn't demand action.

## How it's built

1. Twico pulls every observed inbound Slack message + every outgoing reply activity since you were last focused.
2. Both sides are interleaved chronologically per channel into a real chat transcript: `Sarah (10:34): can you check the API doc?` … `twico (10:35): yeah, link me the version?`
3. A single LLM call (Claude Haiku by default) receives those transcripts and writes:
   * An **overview** — one or two sentences capturing the feel of the window ("Busy hour…" / "Quiet morning…").
   * A **per-conversation summary** — 1–2 sentences per channel, with a `needs_followup: true` flag when the contact is still waiting on you.
4. The dashboard renders those, with follow-up channels highlighted in amber.

## Fallback

If no LLM key is configured (or the call fails), the digest falls back to a deterministic one-line counts summary:

> While you were away (2.4h): answered 12 messages, 1 thread needs your follow-up — busiest in #engineering.

…and the per-conversation block stays hidden.

## When does it not show?

* On first launch (no recorded focus marker yet — there's nothing to be "back from").
* When the gap was less than \~60 seconds.
* When literally nothing happened in the window.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://litao.gitbook.io/twico/how-the-twin-behaves/while-you-were-away.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.