feat(queen): --unchecked moves every draft into the body, minus the citations - #4708
Merged
Merged
Conversation
…itations `--write-body` wrote a boundary only where an issue named exactly one path and it was a `.t27`: 44 issues. That left 123 carrying a draft in a COMMENT and nothing in the body - and the body is the only place the Queen reads. A draft nobody moves across is a draft that does nothing. Owner's decision, 2026-09-24: move them. `--unchecked` drops the one-spec guard and keeps the citation rule. A body line is a CLAIM on a file, which a comment is not: a comment can afford to list a `docs/` path with "delete if the work does not touch it" beside it because it reserves nothing, and a document an issue merely quotes is the line most likely to be wrong. Reserving it blocks whoever really owns it. So a `docs/` path is carried into a comment and left out of a body, and an issue whose every path is a citation is skipped rather than given an empty section: `boundaryPathsOf` reads "no section" and "empty section" the same way, so an empty one would look like a boundary only to a person. 122 bodies written. Measured across the ticks after: missingBoundary 565 -> 398, claimed 41 -> 74, refusal None on every tick, active bees 1 -> 12-16 of 20. Closes #4707 Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Contributor
|
📓 NotebookLM Notebook linked to this PR
This notebook contains session context, decisions, and artifacts for this work. |
Contributor
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes #4707
--write-bodywrote a boundary only where an issue named exactly one path and it was a.t27— 44 issues. That left 123 carrying a draft in a comment and nothing in the body, and the body is the only place the Queen reads. A draft nobody moves across is a draft that does nothing.Owner's decision, 2026-09-24: move them.
What the flag does not drop
A body line is a claim on a file, which a comment is not. A comment can afford to list a
docs/path with "delete if the work does not touch it" beside it, because it reserves nothing. A document an issue merely quotes is the line most likely to be wrong, and reserving it blocks whoever really owns it.So
--uncheckeddrops the one-spec guard and keeps the citation rule:specs/x.t27tools/y.pydocs/plan.mdThe last row matters:
boundaryPathsOfreads "no section" and "empty section" the same way, so an empty one would look like a boundary only to a person.Measured
122 bodies written. Alongside it, 46 issues stuck behind an empty branch were re-filed with
refile.py, which releases a dispatch row that spent its ceiling and would otherwise hold its issue for good.missingBoundaryclaimedrefusal"nothing to choose"Noneon every tick🤖 Generated with Claude Code