Introduction
What Wizardful is, and how it fits into your product.
Wizardful is an in-app AI copilot for B2B SaaS. You add it with one script tag, point it at your docs, and your users can type what they want and have your product do it.
The three modes
Wizardful operates at three levels of trust, and you decide how far it goes — per task.
| Mode | What it does | Default |
|---|---|---|
| Ask | Answers from your docs, with a citation | On |
| Show me | Moves a labelled ghost cursor across your real UI and narrates each step, touching nothing | On |
| Do it for me | Completes the task in the live UI | Off |
A fresh install cannot change anything in your product. Autonomy is something you switch on, per task, once you have watched the skill do it correctly.
Why there's no integration work
Wizardful reads your live DOM the way a person reads a screen. Any button you have already shipped is a button it can already use — so there is no action registry to maintain, no callbacks to wire, and no API to expose. Your existing interface is the API.
Skills, and why it gets cheaper
The first time a user asks for something new, the agent works it out live by reading your UI. That takes about 13 seconds. That run is then distilled into a skill — a deterministic, replayable recipe for that task in your app.
Every user after the first gets the skill instead of the agent: it replays in about 0.7 seconds, makes zero AI calls, and costs nothing. If your UI changes underneath it, the skill's own checks stop passing, it falls back to the agent, solves the task against the new UI, and recompiles itself.
Skills are unlimited on every plan.