Accessibility statement

Our design target for this website is the Web Content Accessibility Guidelines (WCAG) 2.1 at Level AA. That target shapes how the pages are built: real headings and landmarks, text that meets AA contrast on our ink-and-bone palette, visible focus on everything you can operate, and no information carried by colour alone.

We have not commissioned an independent audit, so this is a statement of intent and practice rather than a conformance claim. When an audit happens, this page changes.

• Semantic landmarks and a single, ordered heading structure on every page, so screen reader users can move by region and heading.
• Full keyboard operation: navigation, filters, the agent detail dialog, the conversation accordion, the multi-step meeting form and every text field.
• Dialogs and drawers trap focus while open, close on Escape, and return focus to the control that opened them.
• Visible focus indicators — we never remove the focus ring without replacing it with something clearer.
• The advisor answers as live regions with a status message, so a screen reader hears that an answer is arriving and when it is done, rather than sitting in silence.
• Ordinal cues in the conversation (“Question 2 of 4”) so position is announced, not just shown.
• Reduced motion is respected. If your system asks for less motion, the site stops animating and scroll jumps directly instead of gliding.
• Video is click-to-load behind our own labelled facade, so nothing autoplays and no third-party player loads until you choose it.
• Text scales and reflows to 375px without a horizontal scroll or a hidden control.

Answers and the walkthrough are meant to leave the screen. Print and “save as PDF” use a dedicated print stylesheet: navigation, buttons and decorative rules drop away, the text reflows to a readable measure in black on white, and the document keeps its heading structure so the resulting PDF is navigable rather than a picture of a web page. Link destinations and the question being answered are carried into the printed record.

We test in current versions of Chrome, Edge, Firefox and Safari on desktop, and Safari on iOS and Chrome on Android, at widths from 375px up. Accessibility work is checked by keyboard-only navigation, by inspecting the accessibility tree, and with automated checks during development. We use the assistive technology built into those platforms rather than certifying against a specific screen reader version.

• Voice input in the advisor relies on your browser’s own speech recognition. It is not available in every browser, and where it is missing we fall back to typing rather than replicating it.
• Videos are embedded from YouTube. Captions and player controls come from YouTube, and caption quality varies by video; we are working through our own library.
• The agent library shows a large number of images. On a slow connection they load progressively, and each falls back to initials rather than a broken image.
• Some diagrams carry dense detail. Each has a text description nearby, but a few would still be easier to follow with a longer written alternative — that work is in progress.
• The internal administrative tool is not covered by this statement.

If something here blocked you, we want the specific page and what you were trying to do — that is far more useful than a general report. Email hello@autoagentic.ai or use the contact form.

We treat accessibility defects as defects. We will acknowledge what you report, tell you what we found, and fix what we can fix — and where a fix is going to take time, we will offer another way to get the same information in the meantime.