Repository navigation
fix(planning): show read-back, flip condition and numbered confirmation on the interview page #12103
Description
Refs: #7303
Outcome
#7303 held four interview-page items until an eval case failed first (section "Waiting on #7269 (eval-first)"). Those cases have now run under #9794, and the rule slices they produced are closed:
- fix(planning): read back answers that depart from the recommendation before marking them answered #11187 read back departing answers
- fix(planning): close the final restatement with an active confirm-or-name-numbers request #11188 confirm-or-name-numbers final restatement
- fix(planning): state the condition that would flip each recommendation #11190 state the flip condition
The page still lags behind the rules. This slice brings the interview page in line with them:
- Show the agent's read-back of each answer on the page. Today it goes only to the collapsed Log.
- Add a field for the condition that would flip a recommendation. Today the basis sits behind "Why" and "Checked against" in collapsed Details.
- Make final confirmation read "confirm, or the numbers to change", in place of a one-click Confirm with a free-text "Something's off".
- Reconcile page rule R-D ("read-back asks no question") with eval case
interview-readback-dictated-answer, which expects a confirm. Keep one shared rule in a context file, cut R-D to a pointer plus page mechanics, and add a one-click "Confirm reading".
Done when
- Each read-back the agent posts shows beside its answer on the page, outside the collapsed Log, and a page test asserts it.
- Each recommendation has a visible flip-condition field filled from the question data, and a page test asserts it.
- The final confirmation control takes "confirm" or a list of question numbers, and a page test asserts both paths.
- R-D in the page rules is a pointer to one shared rule in
plugins/planning/skills/interview/context/, which agrees withinterview-readback-dictated-answer. The page has a "Confirm reading" control with its own test, because the eval runs without Bash. -
plugins/planningcarries a.changes/fragment, andci-statusis green on the PR.
Out of scope
- The SKILL rule text, which shipped in fix(planning): read back answers that depart from the recommendation before marking them answered #11187, fix(planning): close the final restatement with an active confirm-or-name-numbers request #11188 and fix(planning): state the condition that would flip each recommendation #11190.
- DX-5 and DX-14, which appear met on main, and DX-16.
- UX-3, parked by decision, and UI-14, kept by decision.
Activity
github-actions commentedon Nov 1, 2026
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Medium
Complexity: Medium
Uncertainty: Known
Flags: none
Blocked by: none
Understood: the interview page lags the shipped rules. It should show each read-back beside its answer, add a visible flip-condition field, and take "confirm, or the numbers to change" at final confirmation. It should also replace page rule R-D with a pointer to one shared rule and add a "Confirm reading" control. The done-when items are concrete and testable. I found no duplicate among open issues. #7303 is the parent reference.
Automated triage, applied by the lanes App.
github-actions commentedon Nov 1, 2026
melodic-automation-lanes commentedon Nov 1, 2026
The work-resolve-issue run that
agent-readystarted waited in its slot past a change to its lane workflow files on main, so it stopped before any model turn. The run's report-outcome removed and addedagent-readyback to start a fresh run on main's workflow files. If this run ends the same way, it re-fires the issue again, waiting longer each time.github-actions commentedon Nov 1, 2026
melodic-automation-lanes commentedon Nov 2, 2026
This records a re-fire of an earlier refusal, not a new refusal. The lanes kill switch stopped run 41355177441, the work-resolve-issue run that
agent-readystarted, before it did any work; that run ended at 2026-11-01T22:32:52Z. work-unblock has now removed and addedagent-readyback to start a fresh run. A kill-switch refusal is no fault in this issue or the lanes, so it never counts against later re-fires; after limits.limit-stops-per-issue of them since a person last addedagent-readyor a run last reported, work-unblock waits longer before each one, up to 24 hours.4 remaining items
github-actions commentedon Nov 2, 2026
melodic-automation-lanes commentedon Nov 2, 2026
The work-resolve-issue lane stopped on this issue (
outcome-unread): the run wrote no readable outcome file: it crashed or was cut off, which says nothing about the issue.It keeps
agent-ready, and work-unblock starts it again (retry 1 of 2).In the run's words:
Partial work is on the branch
work-resolve-issue/issue-12103-41379480063-1. The next run continues from it while it is not behind the default branch, and otherwise starts fresh.Run: https://github.com/melodic-software/claude-code-plugins/actions/runs/41379480063
github-actions commentedon Nov 2, 2026
github-actions commentedon Nov 2, 2026
melodic-automation-lanes commentedon Nov 2, 2026
This records a re-fire of an earlier refusal, not a new refusal. The lanes kill switch stopped run 41385204205, the work-resolve-issue run that
agent-readystarted, before it did any work; that run ended at 2026-11-02T04:40:39Z. work-unblock has now removed and addedagent-readyback to start a fresh run. A kill-switch refusal is no fault in this issue or the lanes, so it never counts against later re-fires; after limits.limit-stops-per-issue of them since a person last addedagent-readyor a run last reported, work-unblock waits longer before each one, up to 24 hours.github-actions commentedon Nov 2, 2026
github-actions commentedon Nov 2, 2026
melodic-automation-lanes commentedon Nov 2, 2026
This run started fresh from the default branch instead of resuming
work-resolve-issue/issue-12103-41379480063-1, which is behind it. That branch keeps the earlier run's commits until the next run on this issue deletes it, when no pull request uses it.