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

# Investigations

> Start an investigation, ask about an incident, or explore your system using natural language.

The Investigations page is where you interact with Traversal's AI SRE. You can start an investigation, ask about an active incident, or explore system behavior, all in natural language.

## Starting an investigation

Type your question or paste an incident description directly into the input. Traversal parses the text for entities (teams, services) and extracts the date and time if present.

### Pasting incident descriptions

You can paste a snippet from a ticket, Slack thread, or alert directly. Traversal extracts the relevant entities and time window automatically.

**Time formats that parse well:**

* `Oct 23 20:00 UTC`
* `10/23/2024 at 20:00 UTC`
* `Oct 23 4pm ET 2024`
* `5pm EST yesterday`

> [!WARNING]
>
> Root cause investigations need a start date and time with a timezone. If you omit the time window, Traversal can't run one.

## Investigation depth

Traversal picks how deeply to investigate based on your request, so you don't need to choose. The modes are:

| Mode | What it does |
| - | - |
| `auto` | Traversal classifies the request and picks the depth. This is the default. |
| `deep` | A thorough root cause analysis. |
| `fast` | A quick, exploratory pass. |
| `lightning` | A quick first pass for agent-to-agent work over [MCP](https://docs.traversal.com/using-traversal/mcp.md#agent-to-agent-mode). |

In the web app, Traversal chooses for you. Over the [Sessions API](https://docs.traversal.com/api/sessions.md#investigation-depth), you can set `auto`, `deep`, or `fast` with `thinking_mode`. MCP calls always use `lightning`.

### Writing effective descriptions

Better descriptions produce better results. Include:

* When the issue started
* Error messages you're seeing
* Affected teams or services
* Links to relevant alerts or runbooks

## What Traversal returns

Each investigation returns a root cause analysis (RCA) with the following:

> [card] Root cause summary
>
> A summary statement with a confidence level and supporting evidence.

> [card] Evidence citations
>
> Clickable citations including metrics anomalies, log lines, deployments, code diffs, and topology relationships.

> [card] Timeline
>
> A sequence showing when symptoms began, anomalies were detected, deployments occurred, and dependent services started failing.

> [card] Recommended next steps
>
> Diagnostic suggestions, remediation steps, runbook-style guidance, and suggested team escalations.

### Confidence levels

| Level | Meaning |
| - | - |
| **High** | Strong correlation patterns with consistent evidence across multiple signals. |
| **Medium** | Theory is plausible and likely, but missing full visibility or traces. |
| **Low** | Weak or partial signals. Requires human confirmation. |

> [!NOTE]
>
> Traversal's recommended next steps are suggestions. Traversal only changes something in your systems if you've given it write access and asked it to.

## Asking follow-up questions

You can continue an investigation by asking follow-up questions in the web app or in the Slack thread for that investigation. Examples:

* Drill into a specific symptom
* Slice the timeline to a narrower window
* Request more detail on a particular signal

> [!TIP]
>
> In Slack, reply with `@Traversal` and additional context, for example: `@Traversal focus on payments-api in prod-us-east after 12:05 UTC`.

> [!NOTE]
>
> To change the analysis time window, start a new investigation with the revised time range rather than asking as a follow-up.

## Next steps

* Read [Understanding results](https://docs.traversal.com/using-traversal/understanding-results.md) to see how confidence and evidence work.
* Deploy a [Worker](https://docs.traversal.com/using-traversal/workers.md) so Traversal investigates incidents and alerts as they happen.
* Start investigations from [Slack](https://docs.traversal.com/using-traversal/slack.md), [Microsoft Teams](https://docs.traversal.com/using-traversal/microsoft-teams.md), or an AI client over [MCP](https://docs.traversal.com/using-traversal/mcp.md).
* Add your team's runbooks and conventions in [Knowledge Bank](https://docs.traversal.com/using-traversal/knowledge-bank.md).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.