feat(skill): implement tiered architecture (v3.0.0) - #30
Closed
edithatogo wants to merge 84 commits into
Closed
Conversation
- Add severity classification for AI patterns (Critical/High/Medium/Low) - Add technical literal preservation rules for code, URLs, paths - Add chain-of-thought reasoning examples - Add common over-corrections section (what NOT to do) - Add self-verification checklist - Update process section with more structured approach - Bump version to 2.2.0 Closes #22
- Add track spec, plan, metadata, and index - Add docs/skill-distribution.md and docs/awesome-agent-entry.md (draft) - Add validation script scripts/validate-skill.sh - Add CI workflow .github/workflows/skill-distribution.yml - Update README and add CONTRIBUTING.md reference to validation Closes #25
…kflow - Add ESLint, Prettier, and TypeScript check configs - Auto-fix formatting across repository - Enforce lint:all and tests in skill-distribution workflow
…et status to in_progress
germanosouza49-beep
added a commit
to germanosouza49-beep/humanizerBR
that referenced
this pull request
Mar 22, 2026
Novos padrões: aberturas formulaicas (blader#26), conectivos de IA (blader#27), anglicismos (blader#28), vocabulário aspiracional vago (blader#29), névoa emocional e agência mística (blader#30), problemas de estrutura e substância (blader#31). Expande padrões existentes com novas palavras proibidas e exemplos. Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Owner
|
Closing this PR because it no longer matches the repo's current direction. Thanks for the contribution. |
jooray
added a commit
to jooray/humanizer
that referenced
this pull request
Apr 5, 2026
Incorporate upstream patterns (passive voice, hyphenated word pairs, persuasive authority tropes, signposting, fragmented headers) while preserving our staccato contrast pattern as blader#30. Total: 30 patterns.
3 tasks
duathron
added a commit
to duathron/humanizer-ext
that referenced
this pull request
May 26, 2026
… true_negative schema After the cross-domain writers' meetup (journalist, academic, blogger, marketing copywriter, technical writer, legal writer, Wikipedia AI-cleanup editor, speechwriter, ghostwriter), every previously unscorable case in the EN pattern corpus has been resolved. Schema changes: - Case dataclass gains true_negative: bool = False field - load_pattern_corpus reads the new field from per-case JSON - score_case (run_pattern_eval) handles true_negative cases via Levenshtein edit-ratio check (passes if ratio ≤ 0.10 — skill correctly left input alone); result reported as status="true_negative" alongside scorable status="scored" and unscorable status="unscorable_*" - run() aggregator tracks true_negatives separately from detection rate; per_pattern_summary now includes true_negatives + true_neg_passes counts Corpus changes (32 cases across 28 pattern JSON files): - 21 cases get populated expected_changes (some narrowed per consensus) - 9 cases become true_negative (`expected_changes: []` + `true_negative: true`) - Each updated case carries notes citing the meetup tally (e.g. "6/2/1") and the specific personas whose dissent drove narrowing decisions - 0 cases remain unscorable. Total corpus state: 51 cases = 42 scorable + 9 true_negative + 0 unscorable (was 19 + 0 + 32 pre-meetup) Unanimous-✓ cases (highest-confidence corpus signals): blader#12 emoji bullets, blader#14 sycophancy, blader#18 stacked adj, blader#29 turn0search, blader#30 utm_source=chatgpt, blader#31 placeholder tokens, blader#32 markdown fence + meta-prompt Stale per-case partials cleared (cached scores were computed against the old expected_changes lists; next pattern eval run will re-score from scratch). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
blader
added a commit
that referenced
this pull request
May 27, 2026
- §14: turn em-dash "overuse" into a hard cut (no em or en dashes in the final rewrite), with a replacement ladder and a final scan. Idea from #96. - §21: expand to cover speculative gap-filling ("maintains a low profile," "keeps personal details private") where a model invents filler instead of saying a source is missing. Idea from #111. - New pattern #30, diff-anchored writing: describe the thing as it is, not as a narration of what changed. Idea from #81. Hand-ported lean versions rather than merging the source PRs. 30 patterns total; README and AGENTS.md updated to match. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
duathron
added a commit
to duathron/humanizer-ext
that referenced
this pull request
May 27, 2026
…rt from upstream) Small patch release. Ports pattern blader#30 "Diff-anchored writing" from upstream `blader/humanizer` v2.7.0 into `patterns/_universal.md` as our pattern blader#41 (our blader#30 slot is already taken by Sentence Intensifiers). The pattern flags documentation, comments, or prose that reads as if it is narrating a change rather than describing the thing as it is. "This function was added to replace the previous approach...", "the CLI now uses YAML (previously it used a custom format)..." — these leak a diff frame into prose that should describe current state. Skip clause is explicit: CHANGELOG, release notes, migration guides, PR descriptions, and refactor blog posts are version-scoped by design and should not be flagged. Total pattern count: 40 -> 41. Universal pack now: {6, 14, 15, 17, 18, 19, 25, 26, 29, 38, 39, 40, 41}. Also updates SKILL.md frontmatter description to mention the `career` domain alongside the other five (was added in v3.4.1 but description was not refreshed). Tests: - UNIVERSAL_PATTERN_IDS extended to include 41 in tests/test_skill_structure.py - Domain-existence assertion extended to include "career" - 78/78 pytest pass Also catches up stale plugin manifests that were last touched at v3.2.0: - `.claude-plugin/plugin.json` version 3.2.0 -> 3.4.2; description refreshed - `.claude-plugin/marketplace.json` description bumped to 41 patterns + career No skill behavior change beyond the new pattern. Pattern eval / FP / E2E baselines from v3.4.1 still represent current state for the 40 pre-existing patterns; pattern blader#41 is not currently in the eval corpus and will be covered in a future corpus expansion. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
duathron
added a commit
to duathron/humanizer-ext
that referenced
this pull request
May 28, 2026
878-line German pattern pack mirroring patterns/en.md structure + extending with 5 DE-only patterns (blader#100-blader#104) per maintainer decisions + Task 5 mining consensus. EN-PARALLEL patterns (translated to DE with DE-specific trigger words + before/after examples): blader#1, blader#2, blader#3, blader#4, blader#5, blader#7, blader#8, blader#9, blader#10, blader#11, blader#12, blader#13, blader#16, blader#20, blader#21, blader#22, blader#23, blader#24, blader#27, blader#28, blader#30, blader#31, blader#32, blader#33, blader#34, blader#35, blader#36, blader#37 (28 patterns). DE-only patterns (blader#100-blader#104, no EN equivalent): blader#100 Akademische Rahmen-Floskel — 'im Rahmen der vorliegenden [Arbeit/Studie/Untersuchung/Analyse]' bureaucratic self-reference. Mining-derived (LLR 31.97, 31:0 AI:human). blader#101 Impersonales Reflexiv — '[es/zusammenfassend] lässt sich [sagen/feststellen/festhalten/zeigen]' AI hedge construction. Mining-derived (LLR 93.73 bigram + 33.00 four-gram). blader#102 Konjunktiv II Stacking — 3+ würde/wäre/hätte/könnte forms in close proximity for vague hedging. Per maintainer OQ2. blader#103 Anglizismen-Leakage — denglisch business buzzwords (insight, deliver, leveragen, Pain Points, ganzheitliche Customer Journey). Per maintainer OQ2. blader#104 Nominalstil-Inflation — noun-heavy bureaucratic verbing ('die Durchführung der Analyse' vs 'analysieren'). Per maintainer OQ2. DE PERSONALITY AND SOUL section mirrors EN with DE-appropriate register notes. Critical addition: domain note excludes DE career writing from soul-adding (DE Anschreiben register is formal-modest, opposite of US/UK puffery — adding soul makes them weaker). blader#7 DE AI Vocabulary trigger list: 33 phrases combining mined tokens (darüber hinaus, zusammenfassend, ganzheitlich, vorliegenden, umfassende, darstellt) with manually curated additions (vielfältig, facettenreich, nachhaltig, innovativ, zukunftsweisend, transformativ, ganzheitlich, intuitiv, nahtlos, robust, im Hinblick auf, vor diesem Hintergrund, es ist wichtig zu betonen, zentrale Rolle spielen, etc.) per OQ1. Excluded per OQ3: 6 Wikipedia-context-only DE-only entries flagged by DE Wiki AI-Cleanup project (productivity spikes, citation format, non-existent categories) — not applicable to general prose. Header documents the exclusion so future contributors don't re-add them. Tests: 207 → 211 passes (+4 DE pack tests: existence, expected pattern IDs, PERSONALITY section presence, no overlap with universal pack). Maintainer flagged for future review: blader#11 Elegant Variation — DE synonym system richer than EN, less sharp blader#34 Trailing Emphasis Fragments — less common in DE, signal stronger when present blader#36 Conditional Frame Stacking — overlaps with blader#102 Konjunktiv II; cross-referenced blader#8 Copula Avoidance — 'gilt als' legitimate legal term of art, apply lightly in legal domain blader#13 Passive Voice — DE academic uses passive more heavily than EN; SKIP in academic AND legal domains (will be enforced in Task 7 domains/de_overrides.md) Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
rinoue-rgb
added a commit
to rinoue-rgb/humanizer-ja
that referenced
this pull request
Aug 13, 2026
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
3 participants
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 #29. Implements the SOTA architecture designed in the Source Verification track. Adds 'modules/' directory and refactors SKILL.md for Router-Retriever pattern.