Build a useful free tool: lessons from Reach
https://patrickz.ai/learn/build-a-free-tool-with-reach/

OUTCOME
Define and test a small accessible tool with an ordinary-device fallback.

YOU NEED
A browser and the free Reach demo (linked below). A camera is optional; start with touch or keyboard. Use fictional messages.

DO THIS FIRST
Helpful first: Design an accessible communication prototype
https://patrickz.ai/learn/design-an-accessible-communication-board/

LINKS
Open the free Reach demo: https://perfectz.github.io/reach-aac/
Reach README and source on GitHub: https://github.com/Perfectz/reach-aac

1. Try the simplest route
Open the free Reach demo linked under Before you start. Try a message using touch or click before enabling any experimental input. Identify select, clear, repeat and pause behavior.

2. Inspect setup as part of the product
Explore language, phrase and input choices. Ask how someone would recover from an unsuitable setting. Camera access should be deliberate, with a visible way to stop it.

3. Choose one contribution
Design a small improvement such as clearer instructions or larger controls. Write a before/after acceptance check and use fictional examples when asking a coding agent for help.

4. Test the fallback
Verify the tool remains usable when camera permission is denied or sound is unavailable. Document device-specific limitations and avoid equating a technical test with clinical validation.

PROMPT
Help me scope one accessible improvement to this communication prototype. Observed friction: [evidence]. Desired behavior: [outcome]. Include the smallest change, keyboard/touch fallback, visible status and a test with camera denied. Use fictional data and state what individual usability testing is still required.

EXAMPLE / EXPERIMENT
A hand-pointer demo is not successful merely because the cursor moves. The person must be able to select the intended message, avoid repeats and pause comfortably.

CHECK YOUR RESULT
[ ] A non-camera route works.
[ ] Permission and pause states are understandable.
[ ] The proposed improvement has a user-visible test.

IF IT FAILS
Reach’s camera inputs are experimental and depend on device, positioning and individual ability. Use the Reach README (linked under Before you start) for current setup, and involve the people who will use it and appropriate support when evaluating real use.

MY RESULT / NEXT CHANGE