You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As raised by @sethdailyssw, we currently run this process manually every week. Let’s add an automated Pull Request audit to SSW.Dory.
The flow should be like this:
For each pull request, pull the title and description from GitHub
Run Prompt 1 - Review PR on the title and description (first comment) together. – it will tell you if the pull request is good or needs correction, and the reason
If output = Good, then nothing happens
If output = Needs Correction, then an email is sent. Template
Figure: Use this email template
Run Prompt 2 on the ALL the results – it will generate a summary email
Send that email:
To: Gordon
CC: Adam, Seth
The text was updated successfully, but these errors were encountered:
Your task is to analyze a pull request provided by the user.
Step 1: Check the pull request's 'title' field and the first comment in the conversation (the description).
Every title must have:
What the Pull Request will do
How the Pull Request achieved it
Examples:
PBI title: Product Backlog Item 100359: "Desktop App | Exporting occasionally failed"
Fix desktop app exporting - prevent database concurrent access while exporting
Context - every Pull Request must have one of the below as the source of the change:
an issue url (or tag the ID, e.g. #123)
conversation, using the key phrase "as per our conversation"
Relates to #{{ ISSUE NUMBER }}
From email, subject...
I noticed that...
Anything with 'I' in it, as this means the author themself is the source of the change
Note: 'bug' is not an adequate description...the person who is the source is not mentioned (needs to be 'I' noticed the bug or '{{ NAME }} noticed the bug')
Write your output in the following format:
‘Good/Needs correction - Pull Request #xxx1 by {{ author }}
Reason: {{ if needed }}’
Gather all the information from the reviewed pull requests and write a message in this exact format:
Note: Good means title and description are adequate, Ok means one of them needs correction, Bad means both title and description need correction. Replace the x with actual numbers.
Hi Adam
Here is your weekly Pull Request report.
Done - SSW.Rules.Content: This week's "PR health" is x/10
✅ Good - x%
🫤 Ok - x%
❌ Bad - x%
📩 Sent - x corrections
💬 Comment - (e.g. Most people are still not putting the source of the change. Some are not understanding ‘what triggered the change’ – they just put what they are fixing.)
Done - SSW.Website (content): This week's "PR health" is x/10
✅ Good - x%
🫤 Ok - x%
❌ Bad - x%
📩 Sent - x corrections
💬 Comment - Most PRs have an issue linked, and the issue usually has the source (some more clear than others)
Cc: @adamcogan @GordonBeeming
As raised by @sethdailyssw, we currently run this process manually every week. Let’s add an automated Pull Request audit to SSW.Dory.
The flow should be like this:
Figure: Use this email template
To: Gordon
CC: Adam, Seth
The text was updated successfully, but these errors were encountered: