# Accessibility Review

KeePassGO currently targets keyboard-first desktop use on Linux and Windows.

## What is intentionally supported

- Keyboard focus is explicit for:
  - vault search
  - breadcrumb navigation
  - entry list selection
  - detail/editor fields
- Focus styling is visible and distinct from the unfocused field treatment.
- Common keyboard workflows are covered in-repo by tests for:
  - tab navigation
  - list navigation
  - search focus
  - new-entry focus transitions
- Controls that participate in keyboard navigation have intent-revealing accessibility labels through `accessibilityLabel` in [`ui_accessibility.go`](/home/jjulian/dev/go/src/git.julianfamily.org/keepassgo/ui_accessibility.go).

## Current screen-reader boundary

- Gio does not currently give KeePassGO a full native accessibility tree comparable to mature desktop UI toolkits.
- KeePassGO therefore treats screen-reader support as:
  - label-conscious where the toolkit exposes focusable controls
  - limited where platform assistive APIs are not surfaced by Gio in the same way as native toolkit widgets
- In practice, this means keyboard and focus behavior are first-class and tested, while spoken output quality still depends on Gio/platform limitations outside this repo.

## Current review result

- Linux:
  - keyboard/focus behavior is intentionally supported
  - visible focus states and control naming are present in code
  - full Orca-style semantic verification is not something this repo can assert automatically today
- Windows:
  - the same keyboard/focus behavior and explicit labels are present in-app
  - full UI Automation parity cannot be claimed from inside this codebase without broader Gio support

## What KeePassGO should continue doing

- Keep every major workflow operable without a pointer device.
- Add explicit labels for any new focusable control.
- Preserve visible focus treatment for new form fields, buttons, and dialogs.
- Prefer dialogs and panels that keep keyboard focus predictable.

## What remains toolkit-limited

- Rich screen-reader semantics beyond the control labeling and focus management done in this repository.
- Native assistive-technology parity with toolkits that expose a fuller accessibility object model.