fix(ci): run the pinned-upstream check even when the link check failed - #11
Merged
Merged
Conversation
3 tasks
napetrov
force-pushed
the
fix/sync-check-runs-when-link-check-fails
branch
from
September 3, 2026 16:07
268fb75 to
fec8b20
Compare
napetrov
changed the base branch from
fix/imported-body-dead-link-warns
to
fix/pages-workflow-permissions-and-pins
September 3, 2026 16:07
This was referenced Sep 3, 2026
napetrov
force-pushed
the
fix/pages-workflow-permissions-and-pins
branch
from
September 3, 2026 16:27
7d52116 to
4b327eb
Compare
napetrov
force-pushed
the
fix/sync-check-runs-when-link-check-fails
branch
from
September 3, 2026 16:27
fec8b20 to
77bd4b2
Compare
napetrov
changed the base branch from
fix/pages-workflow-permissions-and-pins
to
fix/imported-body-dead-link-warns
September 3, 2026 16:27
`Imported skills match their pinned upstream` is the step after `Link check` and had no `if: always()`, so it has never run on a commit where a link was dead. `main` has been red on the link check since this morning, and for that whole window the answer to "does what this repository publishes still match what was reviewed?" was not "yes" — it was unknown, because nothing asked. The two checks are independent. One reads URLs inside skill bodies; the other re-fetches each pinned commit and byte-compares the vendored copy against it. The second is the more serious of the pair — a mismatch means either the copy was edited here or upstream moved the commit under the pin, and both mean this repository is publishing something nobody reviewed. Letting the first decide whether the second runs puts that answer in the hands of whoever last reorganised a documentation site. It does run now, and it passes: every import matches its pin. Ordering is left as it is. `Link check` first is the cheaper and more commonly interesting failure, and with `if: always()` the order no longer decides what gets reported. Signed-off-by: Nikolay Petrov <nikolay.a.petrov@intel.com>
xaleryb
changed the base branch from
fix/imported-body-dead-link-warns
to
main
September 4, 2026 01:11
xaleryb
force-pushed
the
fix/sync-check-runs-when-link-check-fails
branch
2 times, most recently
from
September 4, 2026 01:11
77bd4b2 to
c33fbb6
Compare
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.
4 of 5 in a split stack.
mainmain#9 and #12 are independent and both go straight to
main— they fix the two unrelated jobsthat are red there, and neither waits on the other. #10, #11 and #13 sit on #9 only so their
Validate runs are green while
main's link check is failing; GitHub retargets them tomainwhen #9 lands.
mainis failing two independent jobs, so any PR fixing one still displays the other.Nothing here introduces the failure it shows:
zizmor— the Pages workflow from #7validate— the dead docs.vllm.ai linkEverything else is green on every PR. All five test-merge into
maincleanly in any order,and the merged combination passes the full gate plus zizmor and actionlint.
What this changes
Imported skills match their pinned upstreamis the step afterLink checkand has noif: always(), so it has never run on a commit where a link was dead.mainhas beenred on the link check since this morning, and for that entire window the answer to "does
what this repository publishes still match what was reviewed?" was not yes — it was
unknown, because nothing asked.
The two checks are independent. One reads URLs inside skill bodies. The other re-fetches
each pinned commit and byte-compares the vendored copy against it — and it is the more
serious of the pair, because a mismatch means either the copy was edited here or upstream
moved the commit under the pin, and both mean this repository is publishing something
nobody reviewed. Letting the first decide whether the second runs puts that answer in the
hands of whoever last reorganised a documentation site.
It does run now, and it passes — every import matches its pin:
Step ordering is left alone.
Link checkfirst is the cheaper and more commonlyinteresting failure, and with
if: always()the order no longer decides what getsreported.
Checklist
descriptionagainst requests a user would really type — see CONTRIBUTING.md. (no skill text changed)python3 tools/validate_skills.pypasses locally.git commit -s(DCO).