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

# Overview

> How Actrail works and how to get started.

You connect your agent with the SDK. Its actions appear in the
[console](https://app.actrail.ai), where you turn the risky ones into policies.

## How it works

<Steps>
  <Step title="Capture">The SDK records each action — metadata only.</Step>
  <Step title="Detect">Actrail flags risky patterns when a session ends.</Step>
  <Step title="Decide">You review a flag and turn it into a policy.</Step>
  <Step title="Enforce">The policy blocks the matching action on the agent's next run.</Step>
</Steps>

## Key terms

<CardGroup cols={2}>
  <Card title="Trail" icon="timeline">One agent session — the timeline of what it did.</Card>
  <Card title="Policy" icon="gavel">A rule that allows, flags, or blocks an action.</Card>
  <Card title="Shadow" icon="eye">A policy that reports what it *would* block, without blocking.</Card>
  <Card title="Enforce" icon="shield-halved">A policy that actually blocks the matching action.</Card>
</CardGroup>

## Get started

<CardGroup cols={2}>
  <Card title="Connect Claude Code" icon="rocket" href="/quickstart">
    Set up in about five minutes.
  </Card>

  <Card title="Using another agent?" icon="envelope" href="mailto:team@actrail.ai">
    Only Claude Code is supported in beta — contact the team for early access.
  </Card>
</CardGroup>