Skip to content
BrandWater AI
AgentComing in V3

Actions and Approvals

How do approved changes get made safely?

What you will see

Example finding

Approve a page change

A clarified product page is approved by a named person, executed, and logged with a reversible change record.

Quick answer: Actions and Approvals

Actions and Approvals is the governed execution layer: an approval center, an action library and policies, so that every external change is approved and logged. It is coming in V3.

Coming in V3. Arrives with the third release. This page describes the intended capability so you can see where the product is going. It is not available today.

What you see

An approval center, an action library and agent policies. Every external change needs explicit approval unless a policy you set allows auto execution for a low risk class.

Approval center

Approve, reject or edit each proposed action.

Action library

Content, technical, schema, metadata, internal link, product info, location and citation actions.

Policies

Observe, recommend, execute with approval, or auto execute approved classes.

Guardrails

Domain, URL, class and permission checks that block unsafe actions.

How it works

  1. 1

    Propose

    An action is prepared with its payload and scope.

  2. 2

    Approve

    A permitted person approves, edits or rejects it.

  3. 3

    Execute

    Through a connected CMS or technical route, within policy, and logged.

How it is measured

Definitions come first, so numbers stay comparable from one audit to the next.

Approval record
Who approved, when, the payload and the scope.
Policy level
The permission tier for the workspace or brand.
Guardrail events
Blocked actions, logged with the reason.

Good to know. Coming in V3, starting with one CMS chosen from early customer demand. Nothing executes without approval and policy, and rollback works only where the target system supports it.

Common questions

Will anything change my site without approval?

No action executes outside policy and approval. Auto execution is limited to classes you have explicitly approved.

Which CMS is supported first?

One will be selected from early customer demand. It is not decided here.

Will it execute without approval?

Only for low risk classes you have explicitly approved by policy.

A short checklist

  • Set the policy level
  • Review each payload
  • Approve or edit
  • Confirm the scope
  • Keep the log

Who uses it

Marketing lead
Approve with confidence.
Compliance
Policy and audit trail.
Web team
Clear payloads to apply.

Common mistakes

  • Auto executing too widely

    Start with approval on every action.

  • Skipping scope checks

    Guardrails exist for a reason.

  • Expecting it at launch

    It is coming in V3.

See Actions and Approvals on your own brand.

Start with a first audit. We show where you appear, who is named instead, and what to fix first.