---
title: "After a session"
description: "Review an ended Alpacon work session. Replay its activity timeline and read AI analysis of risk level, summary, detected threats, and recommended actions."
---

> Full documentation index: https://docs.alpacax.com/llms.txt

# After a session

Review a work session after it ends, including its activity and analysis.

## Session review tabs

Everything a work session does is recorded for review on the session's detail page. The **History** tab replays a timeline of the session's activity, and the **Analysis** tab adds an AI risk assessment once the session ends.

## Timeline (History)

The **History** tab shows a chronological **Timeline** of the session's events—who did what, when, and on which server. Each entry is a recorded action:

| Event | What it records |
| --- | --- |
| **Command** | A command run on a server |
| **Web terminal** | A web terminal opened in the browser or with the CLI |
| **Port forwarding** | A port forwarding tunnel |
| **Editor** | A code editor session |
| **File transfer** | A file transfer session |
| **Upload** | A file uploaded to a server |
| **Download** | A file downloaded from a server |
| **Privilege elevation** | A request to run a command with sudo |

### Browsing the timeline

Filter the timeline by category: **All**, **Web terminal**, **File transfer**, **Command**, or **Privilege elevation**.

Long sessions load in pages. If loading stopped partway, a **Partial** label with **Retry** appears at the top of the list.

### Event details

Select an event to see its details. An event shows the **Server** it ran on and the account it ran as (**Run as**) when it has them. The other details depend on the event type:

- **Command**: **Exit code**, **Verification**, **Channel** (**Websh**, **API token**, or **Service token**), and **Shell**.
- **Web terminal**, **Port forwarding**, **Editor**, and **File transfer**: when it was **Opened** and **Closed**, and how long it stayed open (**Open for**). A web terminal and port forwarding also show the **Client** (web or CLI), and port forwarding shows the **Target port**.
- **Upload** and **Download**: **File name**, **Path**, and **Size**.
- **Privilege elevation**: **Allowed by**, whether the grant is **One-time**, the **Approver** when a person approved it, and **Authorized at** and **Expires at** once the grant is in effect.

### Terminal recordings

Select a **Web terminal** event to play its **terminal recording** in place. The recording loads when you open the event, so a spinner shows while it is read. Recordings are **masked** to hide sensitive data.

- If the recording could not be read in full, a **Partial** label appears; select **Retry** to read it again.
- A terminal that finished without printing anything shows that there is no recording.
- While a terminal is still open in a live session, its recording keeps up as new output arrives. If it has not printed anything yet, the event shows that there is no recording yet.

Playing a recording is available on a paid plan. On the Free plan, the event shows **Terminal recordings start on Essentials**, with an action to upgrade, start a trial, or ask a workspace admin, depending on your billing permissions. Commands and the other events are not affected. See [Websh limits](/use/work-sessions/terminal/sessions/#limits) for replay availability and plan information.

From the timeline you can also jump straight to the AI analysis with **View AI analysis**.

## AI analysis

The **Analysis** tab provides an AI review of the session. It becomes available **once the session has ended** (Completed, Expired, or Revoked); before that, the tab notes that analysis isn't available yet.

If a session hasn't been analyzed, click **Request analysis**. While it runs, the tab shows **AI is analyzing the session...** and updates automatically when it's done.

A completed analysis shows:

- **Risk level**: Low risk, Medium risk, High risk, or Critical risk.
- **Summary**: a plain-language overview of what happened.
- **Key metrics**: **Commands** run, **Risk factors** found, **Confidence** of the assessment, and whether an **Attack chain** was **Detected** or **None**.

It then breaks the review into three sections:

- **Activity**: the **executed commands** and a **timeline analysis** of how the session unfolded.
- **Threat analysis**: when anything is found, an **attack chain** view, **threat indicators** (Malware tools, Credential access, Network artifacts, Sensitive file access), and a **MITRE ATT&CK** technique list. If nothing is found, it shows **No threats detected**.
- **Response guide**: **recommended actions** and a **verification guide**. The guide shows whether it's Safe to proceed, then lists manual checks, each with **How** to check it and the **Expected result**.

## Related docs

- [Use a session](/use/work-sessions/detail/)
- [Live activity](/use/live-activity/)
- [Session history](/use/audit/sessions/)
- [Websh session analysis](/use/audit/events/websh/analysis/)

Source: https://docs.alpacax.com/use/audit/sessions/review/
