Skip to main content
You connect your agent with the SDK. Its actions appear in the console, where you turn the risky ones into policies.

How it works

1

Capture

The SDK records each action β€” metadata only.
2

Detect

Actrail flags risky patterns when a session ends.
3

Decide

You review a flag and turn it into a policy.
4

Enforce

The policy blocks the matching action on the agent’s next run.

Key terms

Trail

One agent session β€” the timeline of what it did.

Policy

A rule that allows, flags, or blocks an action.

Shadow

A policy that reports what it would block, without blocking.

Enforce

A policy that actually blocks the matching action.

Get started

Connect Claude Code

Set up in about five minutes.

Using another agent?

Only Claude Code is supported in beta β€” contact the team for early access.