- TODO: Fill in the README sections below the
<hr>
at the bottom of this page. - Github settings:
- TODO: Under "General", turn off github features you don't need, and turn on others that aren't the default. Suggestions:
- "Wikis" off.
- "Projects" off.
- "Allow merge commits" and "Allow rebase merging" off; "Allow squash merging" on.
- "Automatically delete head branches" on.
- TODO: Under "Collaborators and teams" add users besides yourself.
- TODO: Decide on a branch and PR strategy.
- Feature branches from
main
work for many repos, but if a different strategy (like gitflow) would work better, explain the strategy in the README, and update the default branch. - Add branch protection rules. Suggested rules:
- "Require a pull request before merging"
- "Require status checks to pass before merging"
- Feature branches from
- TODO: Under "Code security and analysis", turn on:
- Dependabot
- Code scanning
- TODO: Under "General", turn off github features you don't need, and turn on others that aren't the default. Suggestions:
- Github issues:
- TODO: Decide how you want issues reported, and document that decision in the README. Suggest that you create a
triage
label, and use the provided issue template that applies it automatically. Asking that users go through the helpdesk or slack is also reasonable, as is a commitment to keep an eye open for any incoming issues. - TODO: Establish responsibility for incoming issues. Suggest that one person saves the url for a filtered list of
triage
issues, reviews the list every day, and removes thetriage
label once they have been reviewed. - TODO: Review the issue templates in this repo and edit to fit your needs.
- TODO: Remove issue labels that do not add much value. Suggest removing
duplicate
,good first issue
,help wanted
,invalid
,question
. - TODO: Add issue labels as needed. Suggest adding
triage
.
- TODO: Decide how you want issues reported, and document that decision in the README. Suggest that you create a
- Github CI:
- TODO: Under
.github/workflows
, update the provided examples as appropriate for your toolchain. Suggest starting with linting and tests. Suggest having each check as a separate workflow, so if there is a failure, its easy to tell at a glance what it was.
- TODO: Under