feat(patterns): #34 False Agency across four languages (+ en #9 serial-negation)#618
Merged
Conversation
…a + en #9 serial-negation extension Adopted from reviewing hardikpandya/stop-slop (13.7k★): its one substantive gap in our catalog. Inanimate subjects performing human actions (decisions emerge, cultures shift, data tells) hide the actor in ACTIVE voice, which passive-voice detection (#26/#27) cannot see — and it dovetails with Study 1's finding that surviving human-tells are actor-ed specificity. HIGH semantic risk: rewrite must never fabricate an actor the source does not support; academic metonymy (the study finds), agentless processes, and software subjects are excluded. Also: en #9 now fires on serial-negation build-ups (It wasn't X. It wasn't Y. It was Z.), lexicon re-mining candidates queued (docs/benchmarks/ lexicon-candidates.md — gate-tested, never direct-added), success+failure examples ×4 langs, PATTERNS docs/README/scoring tables synced (172 patterns). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Contributor
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Research outcome from reviewing hardikpandya/stop-slop (13.7k★, en-only prompt skill): most of its catalog is already covered by our 33 patterns with calibration it lacks — the one substantive gap was false agency.
New pattern #34 — False Agency (inanimate actors) — ko/en/zh/ja
"The data tells us / decisions emerged / 문화가 바뀌었다 / 決定が自然と生まれた" — abstractions promoted to actors so the prose sounds analytical while never committing to who did what. Distinct from passive-voice detection (#26/#27): these are active-voice sentences, so passive detectors miss them. Fires on accumulation (2+ with a recoverable-but-unnamed actor); academic metonymy ("the study finds"), genuinely agentless processes, software subjects, and sustained literary personification are excluded. HIGH semantic risk with an explicit no-fabricated-actors preservation rule — same contract as the Study-1-derived specificity guidance.
Also in this PR
docs/benchmarks/lexicon-candidates.md: stop-slop phrases queued for the next lexicon re-mining — candidates must pass the pre-registered corpus gate, never direct-added; blanket bans (all adverbs, no em dashes, no Wh-starters) rejected as style opinions.scripts/research/ko-doc-generate.mjsline 1.Gates
Full suite 1369 pass / 0 fail (three integration tests caught missing doc syncs — fixed), lint clean, benchmark all-fixtures-correct.
🤖 Generated with Claude Code