Rework the PR template and check that it is filled in - #2689
Merged
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
amankrx
requested review from
MarcusSorealheis,
corcillo,
modernmedici and
palfrey
August 13, 2026 09:10
palfrey
approved these changes
Aug 13, 2026
corcillo
approved these changes
Aug 14, 2026
modernmedici
left a comment
Collaborator
There was a problem hiding this comment.
Looks good, just adjusted the wording in one place to make sense.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What and why
The PR template asks for things CI already proves and not much a reviewer actually needs, so it gets filled in as ceremony. Measured across the last 33 merged PRs that used it:
bazel test //...passes locally". Nobody runs the full suite before every PR, so the box carries no information.Several of those are mine. This replaces it with three sections that each ask for something only the author knows: what problem this solves, how it was actually verified, and what breaks if it is wrong. All guidance moved into HTML comments so it cannot leak into the permanent record.
The checklist is gone.
bazel test //..., formatting and lints are already CI jobs and pre-commit hooks, which enforce them properly rather than on the honour system.How this was verified
Ran the checker against real inputs: a filled-in body passes, the unedited template fails on all three sections, a body with no headings fails, an empty body fails, and a body with "n/a" under a heading fails on that heading only. Also ran it against #2683, an existing PR using the old template, and confirmed it reports the headings as missing rather than crashing.
Workflow YAML parses and the triggers resolve as intended. I could not exercise the workflow end to end, since that only happens once it is on the default branch.
Risk
This is a process change, so it needs agreement rather than just review.
The check fails PRs whose description is thin, which is the point, but two details keep it from being hostile:
skip-pr-templatelabel, which a maintainer will need to create.edited, so it goes green as soon as the author fixes it. A red X with no explanation is how you lose a drive-by contributor.Open PRs written against the old template will fail this check the next time they are edited or pushed to. The label is the escape hatch, or hold the merge until the queue drains.
It runs on
pull_request_targetso it can comment on fork PRs, where a first-time contributor most needs the hint. Nothing from the PR is checked out or executed: the checkout is the base branch, and the description is only ever read from a file, never interpolated into a shell command.No new third-party action; it reuses the
actions/checkoutpin already trusted in this repo and the preinstalledghCLI.This change is