Repository navigation
skill: reword one phrase to unbreak the gate my own correction broke - #3251
Merged
Merged
Conversation
`Documented t27c subcommands exist` reads backticked `t27c <sub>` mentions in live documents and fails on any name that does not exist, unless the document says so in words it recognises: not implemented / not built / does not exist / never existed / no such subcommand. #3231 added a paragraph explaining that `t27c gen-zig` is not a subcommand. It says the right thing and matches none of those patterns, so the gate went red. #3231 merged at 02:21:25Z and the gate's first master failure is at 02:21:28Z -- the run its own merge triggered. Seven successes before, five failures after. Worth recording: I then read three red master runs as evidence the gate was pre-existing and not mine, and `tri pr ready` said "also failing in 4 other place(s) -- pre-existing", which is true and does not mean what I took it for. A defect merged to master fails on every later pull request, which is precisely what pre-existing looks like from a branch-versus-master comparison. The separating question is one command: when did this gate last pass, and what landed between. Repair is one phrase. The checker now reports ok. Refs #3241 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
gHashTag
enabled auto-merge (squash)
September 5, 2026 04:31
This was referenced Sep 5, 2026
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
This was referenced Sep 5, 2026
Contributor
This was referenced Sep 5, 2026
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
Contributor
PR DashboardGenerated at: 2026-09-05 05:51:07 UTC
Summary
Seal Status
|
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.
Documented t27c subcommands existhas been red on master since 02:21Z. I put itthere, and when I looked I concluded the opposite.
What happened
#3231 corrected a claim of mine, and its text explains that
t27c gen-zigis not asubcommand — with the name in backticks, which is exactly what the gate reads. The
gate accepts a disclaimer matching
not implemented|not built|does not exist|never existed|no such subcommand|.... My wording was "which is not a subcommand": right inmeaning, matching none of those patterns.
#3231 merged at 02:21:25Z. The gate's first master failure is at 02:21:28Z — the
run its own merge triggered. Seven successes before it, five failures after.
The misreading, which is the part worth keeping
I saw three consecutive red master runs and wrote "this is a pre-existing red gate,
not something my PR caused".
tri pr readyagreed in its own words: "also failing in4 other place(s) — pre-existing".
Both readings are accurate and neither means what I took it to mean. A defect merged
to master fails on every subsequent pull request, which is exactly what "pre-existing"
looks like to a tool that compares a branch against master. The question that
separates them is one command: when did this gate last pass, and what landed between
then and the first failure.
The repair
One phrase — "which is not a subcommand" → "which does not exist as a subcommand".
The checker then reports:
The gate is well built: it names the file, the line, the phrase it wants, and where
the disclaimer may sit — on the line, in the paragraph, or in the heading above it.
Refs #3241