What does WCAG 1.3.5 require?

Success Criterion 1.3.5 Identify Input Purpose requires that fields collecting information about the user expose their purpose programmatically. In HTML that means using the autocomplete attribute with a token from the autofill detail tokens list, such as given-name or email.

Why not just use autocomplete equals name for everything?

Each token has a specific meaning. The token name is the full name, given-name is the first name and family-name is the surname. Using the wrong token makes autofill insert the wrong data, so the value must match what the field actually collects.

Is autocomplete equals off acceptable?

It is allowed when you genuinely need to suppress autofill, for example a one-off confirmation field, but it should be intentional. For fields about the user, turning autocomplete off removes the 1.3.5 benefit, so the tool flags it for review rather than passing it silently.

Does every field need an autocomplete attribute?

No. Only fields that collect information about the user, such as their name, address, phone or payment details, fall under 1.3.5. A free-text search box or a comment field does not need one, so a missing attribute there is not a failure.

Is my markup uploaded anywhere?

No. The HTML is parsed in your browser with the built-in DOM parser and validated locally against the token list. Nothing you paste is sent to a server.

autocomplete Attribute Validator

Autofill is a quiet accessibility feature. When an input carries the correct autocomplete token, browsers and password managers can fill it in automatically, which is a major help for users with motor or cognitive disabilities who would otherwise retype the same data on every form. WCAG 1.3.5 makes this a requirement for fields that collect information about the user, and this tool checks that each token you used is actually valid.

How it works

The tool parses your HTML with the browser’s DOM parser and collects each data-entry input, select and textarea. For every field it reads the autocomplete attribute and validates the value against the grammar from the HTML Living Standard:

The special keywords on and off are recognised directly.
Optional prefixes are stripped in order: a section-* group, then shipping/billing, then a contact type such as home or work.
An optional trailing webauthn token is removed.
The remaining single token is the field name and is checked against the permitted autofill list, for example given-name, family-name, email, tel, street-address, postal-code, cc-number and one-time-code.

When the field name is not in the list, the tool checks a table of common mistakes (such as phone, zip or fullname) and suggests the correct standard token.

Example

<input name="tel" type="tel" autocomplete="phone">

The value phone looks reasonable but is not a real autofill token, so browsers ignore it. The correct token is tel, and the tool reports exactly that substitution.

Tips and notes

Tokens are case-insensitive in HTML, but lower-case is the convention and is easiest to read.
For multi-part fields, prefer the granular tokens: address-line1, address-level2 (city) and address-level1 (state or region) rather than packing everything into one box.
Combine shipping/billing prefixes when a page has both addresses, for example shipping street-address and billing street-address, so autofill keeps them separate.
All validation runs in your browser; the markup you paste never leaves the page.