Jira Acceptance Criteria and Definition of Done Audit
Score a sample of Jira stories for testable acceptance criteria and a shared definition of done, with rewrites and a team checklist.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
[QUESTION FOR PRODUCT OWNER] is an editorial flag, not a blank to fill. Replace or resolve each flag before you use the result.
Please evaluate how well our Jira stories define what finished means. I will paste a sample of stories with their descriptions and acceptance criteria, plus our written definition of done if we have one. Inputs: 8 to 12 sample stories, with key, title, description and criteria [SAMPLE_STORIES], our definition of done text or "none" [DEFINITION_OF_DONE], the kind of product we build [PRODUCT_TYPE], and who tests and accepts work [TESTING_ROLES]. For each story, score from 1 to 5 on: Criteria exist, Criteria are testable (a stranger could check yes or no), Covers the unhappy path or edge cases, Free of design decisions that belong elsewhere, Linked to a user or business need. Put the result in a table with a total out of 25 and a status of Ready, Needs work or Not ready, using cut-offs you state. Show the average per criterion across the sample, with the arithmetic, and identify the weakest. Next, rewrite the two weakest stories' criteria in a Given-When-Then or checklist style, whichever fits, using only information in the sample, and mark gaps as [QUESTION FOR PRODUCT OWNER]. Review our definition of done for clarity and for items that are impossible to verify. Finish with a one-page team checklist for refinement sessions. Guardrails: do not invent product behaviour. Do not edit issues. Keep recommendations realistic for a team that has limited time. Add a worked example with invented scores: a story with scores 4, 2, 1, 5, 3. Show the total 15 out of 25 and the Needs work label.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.
See an example result (fictional data)
Story PAY-41 (invented): "As a customer I can pay by card". Scores: Criteria exist 4, Testable 2, Edge cases 1, No design decisions 5, Linked to need 3. Total 15 / 25. Status: Needs work. Weakness: criteria say "payment works quickly". A tester cannot check "quickly". Rewrite: Given a valid card, when the customer confirms the order, then the order shows Paid and a receipt is emailed. Given a declined card, when the customer confirms, then an error message appears and the basket is unchanged. [QUESTION FOR PRODUCT OWNER] Which card types are in scope? Sample average for Edge cases: (1 + 2 + 1 + 3 + 2) / 5 = 1.8, the weakest area.
Illustrative only, made-up data. Your results will differ and must be reviewed.