> AI agents: this is one page from PostHog's docs. Full index of Markdown docs for LLMs: https://posthog.com/llms.txt

# Getting a robot to watch for you – Session replay pocket guide

[](/pocket-guides.md)Aa

[](/pocket-guides/session-replay/watching-replays.md)

# Getting a robot to watch for you – Session replay pocket guide

You can watch a dozen replays, but you can't watch a thousand. Have a robot watch it for you.

This section covers various automations you can configure to automate replay watching.

## Which robot

Four automations can watch replays on your behalf:

| Robot | What it does | What you get |
| --- | --- | --- |
| Session summary | PostHog AI reads the event stream of one session, or twenty, and writes what happened | A written account, and the pattern across a batch |
| [Replay Vision](/docs/replay-vision.md) [scanner](/docs/replay-vision/scanner-types.md) | Our model that watches every matching replay, continuously | An [observation](/docs/replay-vision/observations.md) per session, saved as a $recording_observed event you can query |
| Session Replay [signal](/docs/self-driving/signals.md) | Spots rage clicks and friction, raises them | A [report](/docs/self-driving/reports.md) in your [inbox](/docs/self-driving/inbox.md), grouped with related findings |
| [Scout](/docs/self-driving/scouts.md) | Queries anything in PostHog on a schedule – events, insights, warehouse tables, and the observations above | A report too, but only when the answer to its question changes |

Only the Replay Vision scanner watches the video. The other three read events, so they know what people did but not what the screen showed them.

## Summarize with PostHog AI

You can ask PostHog AI to find and summarize sessions. PostHog will summarize the both the sessions replay and the events stream containing pageviews, clicks, inputs, and errors recorded during that session.

Rather than watching them all to spot a pattern, ask for the batch and let it surface what's common.

-   [Summarize the last 5 session replays where users dropped off during checkout](https://app.posthog.com/#panel=max:!Summarize%20the%20last%205%20session%20replays%20where%20users%20dropped%20off%20during%20checkout)
-   [What patterns do you see in sessions with rage clicks?](https://app.posthog.com/#panel=max:!What%20patterns%20do%20you%20see%20in%20sessions%20with%20rage%20clicks%3F)
-   [What are users struggling with on the pricing page?](https://app.posthog.com/#panel=max:!What%20are%20users%20struggling%20with%20on%20the%20pricing%20page%3F)
-   [Summarize sessions from enterprise users who churned](https://app.posthog.com/#panel=max:!Summarize%20sessions%20from%20enterprise%20users%20who%20churned)

It's at its best on [experiments](/docs/experiments.md). Metrics tell you which variant won, summaries tell you why.

-   [What behavioral differences do you see between variant A and variant B?](https://app.posthog.com/#panel=max:!What%20behavioral%20differences%20do%20you%20see%20between%20variant%20A%20and%20variant%20B%3F)

Summarizing a session also marks up its player timeline: green segments where things went well, red where they didn't. Click a marker to jump to that moment, or click a segment in the summary sidebar to move the player there. Summarizing a replay makes it easier to navigate for your whole team down the road.

## Set up a scanner

A Replay Vision [scanner](/docs/replay-vision/scanner-types.md) is three things: a **prompt** saying what to look for, a **type** saying what shape the answer takes, and **filters** saying which recordings it runs against. PostHog then asks that question of every matching session, forever.

Every answer is saved as a `$recording_observed` event on the session it came from, along with the model's reasoning – citations included, linking to the moments that convinced it – its confidence, and what the observation cost.

So "how many people got stuck on pricing last week" is a query:

SQL

[Run in PostHog](https://us.posthog.com/sql?open_query=SELECT+count%28%29%0AFROM+events%0AWHERE+event+%3D+'%24recording_observed'%0A++AND+properties.scanner_name+%3D+'Dead-end+pages'%0A++AND+properties.scanner_output_verdict+%3D+'yes'%0A++AND+timestamp+%3E+now%28%29+-+INTERVAL+7+DAY)

```sql
SELECT count()
FROM events
WHERE event = '$recording_observed'
  AND properties.scanner_name = 'Dead-end pages'
  AND properties.scanner_output_verdict = 'yes'
  AND timestamp > now() - INTERVAL 7 DAY
```

## Signals and scouts

Signals and scouts make your product [self-driving](/docs/self-driving.md).

A [signal](/docs/self-driving/signals.md) is a finding with its evidence attached. Session Replay raises them two ways: it spots [rage clicks](/docs/session-replay/how-to-watch-recordings.md) and friction on its own, and any scanner can hand over what it finds by flipping **Emit findings as Signals** in the editor. Related signals group into a [report](/docs/self-driving/reports.md) in your [inbox](/docs/self-driving/inbox.md), so one problem arrives as one item rather than three.

A [scout](/docs/self-driving/scouts.md) lets agents monitor those signals and act on them on your behalf – dig into the root cause, then draft a pull request Fig. 1. You stay in control of what ships.

P21Accounts that left onboarding midway can no longer finish it 2Accounts that left onboarding midway can't get past step two. The workspace they created last time blocks them, and the name it holds is their own3Scout · dead-end pages4ReadyActionable55h ago

ArchiveReview6

Fig. 1 – What a scout files when replay signals add up to something: the claim on the card, the recordings inside the report.Hover over the figure to learn about each element.

You can learn more about scouts in the [self-driving pocket guide](/pocket-guides/self-driving/101.md). You can read more about the report shown above in [The page people can't leave](/pocket-guides/self-driving/pages-people-cant-leave.md).

## Scanner templates

Here are some templates to give you inspiration and get you started.

[Dead endsMonitor“Did the user appear stuck on a page?”Every session comes back yes or no · `scanner_output_verdict`](https://app.posthog.com/replay-vision/new/details?template=dead_end)[User intentClassifier“What was the user trying to do?”One tag: browsing, purchasing, support\_seeking · `scanner_output_tags`](https://app.posthog.com/replay-vision/new/details?template=user_intent)[Session outcomeClassifier“How did this session end?”One tag: task\_completed, task\_abandoned, blocked\_by\_error · `scanner_output_tags`](https://app.posthog.com/replay-vision/new/details?template=session_outcome)[Frustration scoreScorer“How much friction did they hit?”A number on a scale you set · `scanner_output_score`](https://app.posthog.com/replay-vision/new/details?template=frustration_score)[Session summarySummarizer“What happened in this session?”A title and a short narrative · `scanner_output_title, scanner_output_summary`](https://app.posthog.com/replay-vision/new/details?template=session_summary)

More: [Find replays with PostHog AI](/docs/session-replay/find-replays-ai.md) · [Get started](/docs/replay-vision/start-here.md) · [Creating scanners](/docs/replay-vision/creating-scanners.md) · [Scanner types](/docs/replay-vision/scanner-types.md) · [Observations](/docs/replay-vision/observations.md) · [Calibration](/docs/replay-vision/calibration.md) · [Digests and alerts](/docs/replay-vision/actions.md) · [Pocket guide to self-driving](/pocket-guides/self-driving/101.md) · [Signal sources](/docs/self-driving/inbox/sources.md) · [Scouts](/docs/self-driving/scouts.md)

[‹ Watching replays yourself](/pocket-guides/session-replay/watching-replays.md)[All guides](/pocket-guides.md)p. 7 of 7