What every report covers 7 test types
Seven independent test types, one URL, one report.
Every audit runs the same battery: a visual + functional analysis, a WCAG-mapped accessibility audit, generated test flows, an exploratory agent, persona-perspective reviews, deduplicated multi-page coverage, and any custom checks you've defined. Each is described in detail below — what it does, what it finds, and what it honestly doesn't cover.
The seven types
Foundation
Page Analysis
Vision-LLM audit of a full-page screenshot + DOM + console + network. Surfaces visual, usability, performance, reliability, security, and content issues.
Compliance
Accessibility / WCAG 2.2
In-browser audit across structure, ARIA, forms, keyboard, and contrast. Per-criterion pass/fail, A / AA / AAA grades, VPAT export in deep mode.
Functional
Test Flows
LLM proposes realistic user journeys (signup, search, checkout) and the runner replays each one step-by-step. Pass/fail with per-step screenshots.
Agentic
Exploratory Agent
A second browser session driven by an LLM that picks one action at a time. Eight test types (smoke, negative, edge, monkey, adversarial, …) across eight focuses.
UX perspective
Persona Feedback
Four to six generated personas rate the page on usability, design, content, trust, and accessibility. Surfaces friction a single reviewer would miss.
Breadth
Subpages
After the entry page, the LLM picks the most important linked pages (pricing, signup, checkout, help) and runs the same battery on each. Duplicate findings collapsed.
Your rules
Custom Tests
User-defined flows ("a user can sign up, verify email, log in") and rules ("no console errors on checkout") that run every audit.
How they compose
Every report runs all seven (subject to your plan's page-credit budget). The seven types are intentionally redundant in places — a focus-trap bug, for instance, might surface in WCAG (keyboard 2.1.2), Personas ("I couldn't tab past the modal"), and Exploratory ("the dialog wouldn't close"). The report deduplicates and groups these into a single finding with three pieces of evidence, which is much stronger than one source's verdict alone.
Cost is measured in page credits: standard intensity is 1 credit per page; deep intensity (adds the exploratory agent's full step budget + the WCAG-AAA + VPAT path) is 3 credits per page. See pricing for the per-plan allowance.
