feat(doc-slop): support skipping finding categories - #662
Merged
Conversation
Co-authored-by: Codex <noreply@openai.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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this change add?
The question is how a caller can exempt a named finding category from a fixed-format document without changing the rubric. The rubric is the review's category and threshold rules; a fixed-format document is one whose required section names, headings, or verdict structure cannot be freely changed. A finding category is the identifier attached to one review problem. The deterministic tier applies textlint, a tool that flags Markdown prose problems, and regular-expression checks, while the model tier uses a blind first-reader judge.
scripts/doc_slop_review.pynow accepts repeatable--skip-category NAMEflags, whereNAMEis an identifier emitted by the rubric, a deterministic rule, or a model check. Matching findings are suppressed, per-category counts are reported in text and JSON output, and unknown names produce a warning. With no flag, the existing behavior and output remain unchanged.Why does the tool need it?
Literature notes with fixed section names, review comments with required verdict sections, and skill instructions with required command-style headings can conflict with the new rule that asks headings to be questions, even when those structures are useful to readers. The requested resolution is to keep those structures and let each caller opt out of only the conflicting category; the general rubric remains intact.
What can a caller rely on?
A caller can rely on unchanged default output, suppression of only the explicitly named category in both review tiers, complete suppression counts in text and JSON, and a warning for unknown names without changing the review threshold.