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

# Conversations

> Transcripts, summaries, recordings, and on-demand AI analysis for every call your agents handle.

A **conversation** is one end-to-end interaction — a call, a chat thread, a
voicemail. Each conversation belongs to one [agent](/docs/agents) and owns:

* A `transcripts` row with the full turn-by-turn content.
* An optional recording (audio for voice, transcript-only for chat).
* Zero or more `conversation_analyses` — the assessment saved each time the call
  is analyzed.
* Zero or more `analysis_findings` — per-call observations, each tagged with a
  [finding type](/docs/findings).

Every conversation also carries the basics you see in the list: the agent, a
reference id (the provider's call id), the ended reason, duration, cost, and
whether it arrived from a provider webhook or a direct upload.

## Summaries

The conversation detail page can generate a short AI summary on demand. Click
**Generate summary** and Zelto summarizes the transcript and caches the result
on the row; opening the call again reads the cached summary instead of
regenerating it.

## Filtering

The conversations list filters on:

| Filter              | Matches                                              |
| ------------------- | ---------------------------------------------------- |
| Agent               | Calls handled by one [agent](/docs/agents).          |
| Name / Reference ID | A conversation's title, or the provider's call id.   |
| Ended reason        | The call's disposition (e.g. `customer-ended-call`). |
| Has transcript      | Conversations that have transcript content.          |
| Has audio           | Conversations with a playable recording.             |
| Date range          | Filtered against when the call was ingested.         |

Filters persist in the URL, so a filtered list is shareable.

## Findings and buckets

Any conversation can be linked to a [finding](/docs/findings) — the durable,
shareable issue it's an example of — and curated findings can be pushed to
[Linear](/docs/integrations/linear) from the finding's panel. You can also group
related calls into a static [bucket](/docs/buckets) to analyze or hand off as a
set. The assessment saved when a call is analyzed lives in its
`conversation_analyses` row.