Respecify the out-of-scope dimension as a scope-boundary row - #17
Open
Carr1005 wants to merge 1 commit into
Open
Respecify the out-of-scope dimension as a scope-boundary row#17Carr1005 wants to merge 1 commit into
Carr1005 wants to merge 1 commit into
Conversation
Every spec this guide generates inherits the same defect, found in the agent-memory course spec: the out-of-scope row asks "anything explicitly unwanted?" without ever showing the learner what is already excluded. The exclusion list lives in §1, the §0 gate presents only Ledger rows, and the row's default is a pointer at a section the gate never renders. The dimension is also near-empty as written. Subtraction belongs to integration mode, which this guide does not target; in standalone- takeaway mode the deliverable is fixed and pinned by the acceptance criteria, and every plausible subtraction is already an Options pick on another row. Respecified as a scope-boundary row: the question carries the §1 list, the default is keep-as-is, the options offer restoration without naming what the learner should want back, and each excluded item gets a handling rule (restorable / buildable-but-unsupported / defer to the owning row / unavailable in this build mode). Adds a §14 pre-handoff checklist line so the generating agent catches a non-compliant row before handoff. The §6.0 gate template is deliberately untouched — editing it risks weakening the one-question-per-row and hard-stop rules that §12.10 and §12.11 exist to protect. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The upstream fix for #15 and #16. Those patch two existing specs; this one makes future specs come out right.
The problem
Every spec this guide generates inherits the same defect:
The agent-memory spec shows where that lands: an Options cell reading
Add exclusions freely— one entry, so the row can't even be asked through a structured question tool.The change
Respecifies the dimension as a scope-boundary row (§5.5), with four requirements:
Plus a §14 pre-handoff checklist line, so the generating agent catches a non-compliant row before you ever see the file, and a note on the §1 template that its bullets get read aloud in a question.
What is deliberately not changed
The §6.0 gate template. It would be natural to add a clause there licensing the follow-up that captures which items to restore. I left it alone: that template is copied verbatim into every generated spec, and edits near the one-question-per-row and hard-stop rules risk re-opening the exact failures §12.10 and §12.11 document. The row spec carries the requirement instead.
Diff shape
+17 / -5, all in §3, §5.5, §6, and §14. Four are the dimension rename (out-of-scope→scope-boundary) at its naming sites; the rest is the new §5.5 paragraph, the §1 template note, and the checklist line. Verified the §6.0 template has zero changed lines.Review order
#15 and #16 are independent of this and can merge in any order — they hand-apply what this guide would now generate. If you'd rather see the generated output first, merge those two and re-run
/generate-specon a course after this one lands to confirm a fresh spec produces the same row.🤖 Generated with Claude Code