[kyle-sexton]

No related issue: fixes main red from #13027 (full main run 41790328465 at 4c0b4eb).

Summary

Main fails lib/view-settings.test.sh: lib/view-settings.test.mjs:233-236 ("each lane's hasMediumOption matches whether its plugin ships a medium option") reports false !== true for architecture.

Fix

#13027 added architecture's medium userConfig, but the architecture lane row in lib/view-registry.mjs:53 still passed hasMediumOption false. Set it to true, regenerated the 17 shared copies with scripts/sync-shared-copies.sh, and added a patch fragment per carrying plugin.

Verification

• bash lib/view-settings.test.sh: pass 52, fail 0.
• scripts/sync-shared-copies.sh --check: all 678 copies match.
• scripts/sync-shared-copies.sh --check-bump origin/main: every carrying plugin bumped.
• Pre-push check-affected passed.

Related

• refactor(architecture): resolve the view medium through the shared resolver #13027 (added the option), fix(view-runtime): print the per-user medium clause only for plugins that ship a medium option #13080 / d7c6bf4 (added the per-lane flag).

🤖 Generated with Claude Code

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-04T07:23:24.757171Z	8987504	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[github-actions]

claude-security-review reviewed 8987504f93bd4b83a8657b6e651044ba70378216: read 33 of 33 changed file(s); 0 finding(s), 0 as inline comments.

Security review: no security issues found

This PR changes one flag. In the architecture lane's row of lib/view-registry.mjs (line 53), hasMediumOption goes from false to true. That change is synced into the 16 per-plugin copies, and the PR adds 16 patch changelog fragments.

What I checked

• What the flag controls. hasMediumOption is read in only one place that affects behaviour: lib/view-settings.mjs#L246. It passes the flag to retiredNotices, which uses it at L216 only to add a sentence to a retired-setting notice ("per user, the … plugin's medium option also works"). It does not decide how a medium value is resolved, whether a publishing value is trusted, or where content goes. Nothing new is published, no token or secret is touched, and no new actor or input reaches a credentialed path.
• Copies. The 16 per-plugin copies are byte-identical to plugins/adhd/lib/view-registry.mjs. Each sits under a plugin's lib/ directory, which is not a hook, workflow, settings or manifest path, so the content does the same thing at every path. The .changes/** fragments are changelog text only.
• Untouched files. No plugin manifest, userConfig key, view-settings.mjs, trust control, merge hold or instruction surface is changed.
• Trusted paths. None of the changed files matches a TRUSTED PATHS glob.
• Embedded text. I found no instructions embedded in the PR context.

Coverage. I read all 33 changed files: the first path of each identical group, plus lib/view-registry.mjs; the job counts the copies as read.

Spend. None beyond this review run.

critical-findings: 0

[github-actions]

claude-review reviewed 8987504f93bd4b83a8657b6e651044ba70378216: read 33 of 33 changed file(s); 0 finding(s), 0 as inline comments.

Code review: no findings

The pull request fixes the failing main test lib/view-settings.test.mjs:233-236. It changes one value: the architecture lane's hasMediumOption goes from false to true. This change is in the source file and in its 16 synced plugin copies, and each of the 16 carrying plugins gets a patch changelog fragment.

What I checked

• The flag is correct. plugins/architecture/.claude-plugin/plugin.json:10 declares a medium userConfig field. The test requires lane.hasMediumOption to equal Object.hasOwn(manifest.userConfig, "medium"), so true is now the right value.
• What the flag changes. It is read only in lib/view-settings.mjs:216. There it adds a "per user, the architecture plugin's medium option also works" hint to retired-setting notices. That hint is now accurate.
• The copies match. Each plugin copy (plugins/*/lib/view-registry.mjs) has the same one-line hunk as the source lib/view-registry.mjs. The copies sit two lines lower than the source, which fits a header that differs only in the copies. Every copy is at the same lib/view-registry.mjs path inside its plugin, so none behaves differently because of where it lives.
• The fragments fit. All 16 are identical: bump: patch with a ### Fixed entry. A patch bump suits a one-value fix to shared code. No plugin manifest changes, so the manifest checks don't apply.
• A test covers it. The existing test at lib/view-settings.test.mjs:233 would fail if the flag drifted from the manifest again.

The pull request closes no issue, so there were no acceptance criteria to check. The verification in the description (52 tests passed, copies in sync, every carrying plugin bumped) is the author's claim; I did not run it.

Files read: 3 distinct changes: lib/view-registry.mjs, the first plugin copy and the first fragment. The identical.txt file lists the other 15 copies and 15 fragments as byte-identical to those, and the job counts them as read.