Prompt Almanac

Audit intermediate Works with: Any AI

Azure DevOps Code Review Process Audit

Table output Has example

Check review speed, size and quality of pull requests.

Open in ChatGPTOpen in Claude

The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.

Rate this prompt

Loading ratings…

Audit the pull request process with this data: [PR_DATA with size, time to first review, time to merge, number of reviewers, comments, rework]. Our policy: [POLICY]. Size limit in lines: [SIZE_LIMIT].

Give the sample size and date range. Calculate the median and 90th percentile for time to first review and time to merge (median rather than average where data is skewed), the share of pull requests over the size limit, the share merged without review comments, and the share with rework. Report small, medium and large pull requests separately. Show the arithmetic in a table with the measure, value, sample size and what it means in practice, for example a 90th percentile of 3 days means one review in ten waits longer than 3 days.

Note unusually fast reviews of large changes as a question for the team, not an accusation, and do not rank individuals. Give three hypotheses for slow reviews with a cheap test for each, and suggest working agreements: review within one working day, a maximum size, and how to handle urgent fixes.

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)
Invented sample of 40 pull requests. Time to first review: median 5 hours, 90th percentile 26 hours. Time to merge: median 1.4 days. Over 400 changed lines: 9 of 40 = 22.5 percent. Merged with no comments: 14 of 40 = 35 percent.
Hypothesis: large pull requests wait longer. Test: compare median wait for large and small ones. Reading: a 90th percentile of 26 hours means one review in ten waits more than a day, which is the group worth examining. Next step: list the pull requests with the longest waits and check whether they are large, from new contributors, or raised late in the day. Caution: with 40 pull requests the percentiles move a lot when one or two are added, so repeat the check next month.

Illustrative only, made-up data. Your results will differ and must be reviewed.

Related prompts