Discover Apps Worth Using.

Review this code change before it ships. The change (diff or new code): {{diff}} What it's for, and facts the code doesn't show (data types, how it's called, the rules it must follow): {{context}} 1. List each problem with the line or function, what goes wrong, and a concrete input or sequence of events that triggers it. Check correctness against the rules I gave, security (injection, access control, secrets), data loss, concurrency and repeated runs, error handling, and wrong assumptions about types, units or time zones. 2. Rank the problems: blocks shipping (wrong results, security hole, data loss), should fix before merge, worth a follow-up. Leave out style preferences. 3. Mark each problem as certain from the code shown, or depending on code or behaviour you can't see. For the second kind, say what to check. 4. Give the smallest fix for each blocking problem, as code. 5. List the tests that would have caught the blocking problems, one line each. Don't report problems in code you weren't shown, and don't call code wrong only because you would have written it differently.

Write test cases for this feature from its requirements. Requirements: {{requirements}} What else I know (platform, account types, limits, existing behaviour): {{context}} 1. Group the cases: main path, limits and boundaries, invalid input, account types and permissions, timing (expiry, repeats, two things at once), and failures of services the feature depends on. 2. For each case give a short name, the starting state, the steps, and the expected result taken from the requirements. For every limit, test just inside, exactly at, and just outside it. 3. Where the requirements don't say what should happen, don't choose an answer: write the expected result as "Not specified" and list the question at the end. 4. Mark the cases to run first: the ones that would hurt most if they failed (security, data loss, money) and the main path. 5. End with the open questions, each naming the case it blocks. Use only the rules I gave. Don't invent limits, messages or behaviour.

Write the interface copy for every state of this screen. The screen and what users do on it: {{screen}} Product voice and any length limits: {{voice}} 1. First list the states this screen can be in: first-time empty, loading, each error that can happen, success, and any permission or plan limit. Add a state I didn't mention only when what I described implies it, and mark it as added. Don't add features I didn't describe, such as search, filters or anything that happens automatically. 2. For each state write a heading if it needs one, one or two lines of body text, and the button or link text. 3. Every error says what happened in plain words and what the user can do next. No error codes, no blaming the user, no "Oops". 4. Buttons start with a verb and say what happens ("Save changes", not "Submit"). Use the same word for the same action in every state. 5. Stay within the limits I gave and show the character count next to each line that has a limit. With no limits given, keep headings under 40 characters and body text under 120. 6. End with the terms you used for key things (for example "project", not "workspace") so the rest of the product can match. Use only facts about the product that I gave you. Where a message needs a fact I didn't give (a limit, a time, a contact), put it in [brackets] for me to fill.