[d-morrison]

Moves lms-ref in lint-changed-lines.yaml to psc 1b31160 (psc#89), so the lint on changed lines flags aes() nested in ggplot() and data passed to ggplot() instead of piped.

One line changed; no content.

---

Generated by Claude Code

[github-actions]

Claude finished review — View run

---

Review: #467 — "Bump lms-ref to pick up ggplot_args_linter"

Scope. Single-line change in .github/workflows/lint-changed-lines.yaml, bumping the pinned lms-ref input to the reusable Morrison-Lab/gha/.github/workflows/lint-changed-lines.yml@v3 workflow:

-      lms-ref: 1490eaac059118041bc17e7a846618eac2d41320
+      lms-ref: 1b311601ef3f8226072ec6b660b045f749c916ae

(Per the gha#598 workflow-file fallback note, the on-disk copy under .github/workflows/ in this checkout is the default-branch version, not the PR's; the diff above comes from the saved PR diff file.)

Verification performed:

• Confirmed 1b311601ef3f8226072ec6b660b045f749c916ae is the actual merge commit of Morrison-Lab/psc#89 ("Add ggplot_args_linter: pipe data into ggplot(), aes() as its own layer") and is the current tip of psc's default branch — not a hallucinated or stale SHA.
• The psc#89 description matches this PR's stated rationale: ggplot_args_linter() flags aes() nested inside ggplot() and non-piped data passed to ggplot(), closing a gap where nested_call_linter() alone let ggplot(df, aes(x, y)) pass.
• psc#89 reports 17 passing test expectations and a full lms-suite pass; the linter is included in default_linters(nested_calls = TRUE), which is exactly the configuration this repo's lint-changed-lines.yaml already uses (CLAUDE.md "No nested calls in R" section), so the bump is consistent with, and strengthens, that existing convention rather than introducing a new one.
• No .qmd/prose files are touched, so the SERG-lab-manual prose checks and the LaTeX-macro check don't apply to this diff; no math was added or changed.
• No CLAUDE.md violations: the PR is tightly scoped (one line, no smuggled refactors), the PR description explains the why and cites the upstream PR, and no _site//_freeze/ artifacts are touched.
• No prior Claude comments exist on this PR, so this is the first review round.

Findings: none.

Verdict

Ready for merge — the SHA bump is verified against the actual upstream merge commit, the change is correctly scoped and documented, and no CLAUDE.md or lab-manual issues apply to this one-line CI pin update.

Structured Review Data (JSON)

{
  "schema_version": "1.1",
  "reviewer": "claude",
  "commit_sha": "e910ddef5f8b6d2c8b222ec10837436a672c600b",
  "verdict": "CLEAN",
  "findings": [],
  "detailed_assessment": "The only changed path, .github/workflows/lint-changed-lines.yaml, swaps one verified full-length commit SHA for another with no logic, syntax, or scope issues found.",
  "holistic_assessment": "The new lms-ref matches psc PR 89's actual merge commit and default-branch tip, integrates cleanly with the existing nested_calls=TRUE lint configuration, and needs no further validation beyond CI running green."
}

Reviewed commit: e910dde

[github-actions]

💰 Cost: $0.2416 (review) — run

[github-actions]

PR Preview Action v1.8.1
Preview removed because the pull request was closed.

_{2026-11-03 21:18 UTC}

[d-morrison]

Merged under Ezra's mwc (cmsg_014QLd5McmoUEgMPRiXeVnjSMc2HjrQ1gQkgWs8axE5kgn, 2026-11-03 21:13 UTC). Merge checker: fully clean on head e910dde, annotations read. Cross-harness review waived: the agy and OpenCode Go quotas are exhausted.

Posted by Claude Code (AI agent) --- not written by a human.

---

Generated by Claude Code