Skip to content

feat(planning): add a Later line to the interview round header - #16456

Merged
kyle-sexton merged 2 commits into
mainfrom
feat/planning-interview-later-line
Nov 15, 2026
Merged

kyle-sexton merged 2 commits into
mainfrom
feat/planning-interview-later-line

Conversation

@kyle-sexton

@kyle-sexton kyle-sexton commented Nov 15, 2026 •

Copy link
Copy Markdown
Member

No related issue: an owner decision from a design interview (2026-11-14), recorded outside this repository.

Summary

An interview round said nothing about the rounds that would follow it, so the person answering
could not tell whether a topic was forgotten or only not ready. The round header now carries one
Later: line naming the topics of rounds known to follow.

Fix

  • plugins/planning/skills/interview/SKILL.md: one paragraph beside the Held: and
    Settled by evidence: header lines. The Later: line lists topics, not questions: nothing on
    it takes a Q<N> or a register row, and a topic leaves the line once its questions are asked
    or an answer removes the need for them.
  • plugins/planning/tests/interview-defenses.test.sh: the Stance section digest is re-pinned
    with tests/repin.sh. Justification given to the script: the added line names topics of rounds
    not yet ready; it takes no Q<N> and no register row, asks nothing and resolves nothing, so the
    in-round no-silent-resolve rule is untouched.
  • plugins/planning/skills/interview/context/lane-page.md: the Later: line goes on the page's
    header card.
  • plugins/planning/evals/interview-later-line-known-rounds/: one outcome case. Its grader fails
    a Later: line that carries a question id.

Verification

  • plugins/planning/tests/interview-defenses.test.sh: PASS=185 FAIL=0.
  • claude plugin eval on the new case, 3 runs, with --scaffold: 3 of 3 with this change, 0 of 3 against the plugin as it is on main.
  • scripts/check-affected.sh --tests at this head: 17 passed, 2 skipped.

Remaining

  • The review asked where the Later: line goes under single pacing and on the page. The page
    rule is in this change (context/lane-page.md: the header card). The single-pacing clause
    belongs in the interview skill's Stance section and in context/loop.md, both digest-pinned,
    and needs a second re-pin that was not made here.

Related

🤖 Generated with Claude Code

A round header names the topics of rounds known to follow and not ready yet. It lists topics, not questions: nothing on it takes a Q<N> or a register row. The Stance section digest is re-pinned for the added paragraph.

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@kyle-sexton
kyle-sexton marked this pull request as ready for review November 15, 2026 21:55
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Nov 15, 2026 •

Copy link
Copy Markdown
Codex Review Summary: ✅ Completed

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

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-11-15T21:56:56.073909Z 0387ea8 Draft marked ready
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

Copy link
Copy Markdown
Contributor

Lane run report

Field Value
Activity pr-refine/fix-docs
Model rung 0
Leg handoff
Run job success
Activity step success
Commits pushed none (read lane)
Run 43766403622

Failed GitHub API calls

Method Path Status
GET /repos/{repo}/git/ref/tags/* 404

@github-actions

github-actions Bot commented Nov 15, 2026 •

Copy link
Copy Markdown
Contributor

claude-security-review reviewed f25ae4166a9306e5c9e482f243218b4c6d5d3359: read 3 of 3 changed file(s); 0 finding(s), 0 as inline comments.

Security review: PR #16456, incremental (0387ea8 → f25ae41)

I found no security issues in this push. This lane already reviewed an earlier head, so this pass reported CRITICAL and IMPORTANT findings only. There were none at any severity.

I read every hunk of all 3 files on the incremental list in .claude-lane/incremental.diff. None of them match a trusted-path glob.

What changed

  • plugins/planning/evals/interview-later-line-known-rounds/graders/later-line.md: the grader regex gains a negative lookahead, (?![^\n]*\bQ\d). A Later: line that names question ids now fails. The lookahead scans to the end of the line, the pattern stays anchored to lines, and no nested quantifier can backtrack badly. The grader only scores eval output and gates no merge path.
  • plugins/planning/evals/interview-later-line-known-rounds/samples/later-line.json: one fixed sample is added to the fail list. It is static test data.
  • plugins/planning/skills/interview/context/lane-page.md: one sentence is added, "The Later: line goes on the header card." It only says where an output line is placed. Nothing is deleted, narrowed or softened, so the instruction-surface lens finds no removed rule. It grants no tool, runs no command and touches no gate.

Other checks

  • There are no plugin.json or marketplace.json changes, so there are no manifest lines to report.
  • No trust controls, workflows, tokens or merge holds are touched.

Spend: the push adds no billed resource. The eval case still uses prepaid model credit, as the earlier review noted.

critical-findings: 0

@github-actions

github-actions Bot commented Nov 15, 2026 •

Copy link
Copy Markdown
Contributor

claude-review reviewed f25ae4166a9306e5c9e482f243218b4c6d5d3359: read 3 of 3 changed file(s); 0 finding(s), 0 as inline comments.

Code review (incremental, blocking only): Later: line placement and question-id check

This review covers only what changed since 0387ea8c: three files, and I read every hunk of each. This lane already reviewed an earlier head, so this pass reports only blocking findings. It found none.

What changed

  • graders/later-line.md: the pattern now ends in (?![^\n]*\bQ\d)\S, so a Later: line that carries a question id fails. I traced the samples by hand against the pattern:
    • Both pass samples still match, including the **Later:** bold form.
    • The new fail sample, Later: Q2 (broker), Q3 (schema), is rejected even when [ \t]* backtracks: the lookahead still sees Q2 later on the line.
    • The two older fail samples still have no matching line.
    • Other graders in the repository already use negative lookahead, so the regex grader supports it.
  • context/lane-page.md: adds "The Later: line goes on the header card." This answers the page-surface half of the earlier advisory finding about the Later: line's surfaces.

Not reported
Because this pass is blocking only, two advisory points are left out:

  • The single-pacing half of the earlier surface finding lives in SKILL.md, which is outside this incremental scope.
  • With the i flag, the grader would also fail a Later: line that names a calendar quarter such as Q3.

Not verified: I did not run the eval or the tests. Any results claimed in the PR are the author's and unverified.

Spend: this eval case uses prepaid model usage each time it runs. Not a hold.

Verifier dropped 0 of 0 candidates.

…a Later line that carries question ids

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@kyle-sexton
kyle-sexton added this pull request to the merge queue Nov 15, 2026
Merged via the queue into main with commit f555656 Nov 15, 2026
59 checks passed
@kyle-sexton
kyle-sexton deleted the feat/planning-interview-later-line branch November 15, 2026 22:22
@github-actions github-actions Bot added the merge-path: owner-agent Post-merge accounting record (README.md, Merge paths) label Nov 15, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

merge-path: owner-agent Post-merge accounting record (README.md, Merge paths)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant