Skip to content
LaunchOfferwall

AI assistant

An AI assistant that runs your network with you, not around you

Ask in plain language: how did offers do this week, which conversions are held, add this advertiser's offer. The assistant, powered by Claude, works through the same API you do, with your role's permissions, and asks before it changes anything.

Permissions

It can never do more than you could by clicking

When you open the assistant, it is offered only the tools your role allows, and each tool call runs with your session against the same permission checks and validation as the console. An analyst's assistant can read reports and conversions but cannot change them. A role without write permissions gets a read-only assistant that explains how to make changes instead.

  • Same API, same session, same permission checks
  • Tools filtered by your role before the conversation starts
  • Actions taken through the assistant are recorded in the audit log

Confirmation

Every change waits for your click

Reads run immediately. When the assistant wants to change something, such as creating an offer, reversing a conversion or inviting a teammate, it stops and shows a confirmation card describing the action. Nothing runs until you approve it, and you can decline any card.

Untrusted data

Offer text is data, never instructions

Partner feeds, offer descriptions, postback data and publisher names come from third parties. The assistant is instructed to treat them as untrusted data and never follow instructions found inside them, and the permission and confirmation rules hold regardless of what the text says.

Frequently asked questions

Something else on your mind? hello@launchofferwall.com

Which AI model does it use?

The assistant is powered by Claude, from Anthropic. Requests are sent to Anthropic's API to generate responses.

Is it included in every plan?

Yes. Plans include 500, 5,000 or 50,000 assistant requests per month. When the monthly allowance is used up, the assistant tells you and the rest of the console keeps working.

Will it ask for my partner API keys?

No. The assistant is told never to ask for keys in chat, except when you choose to preview a custom partner template against a live API. You connect partners on the Partners page.

Can I turn it off for some team members?

Using the assistant is a permission (ai.use) that every built-in role has; the assistant always acts within the rest of that member's role.

Try the assistant on your own network