writing-quality

A local detector for AI-writing patterns. Scores every prose file your agent saves. Python standard library only, no network, no model.

Find and repair substance defects in AI-assisted prose, code, docs, and agent output. Reports defects, never authorship. Structural tests over model judgement, because LLM judges agree with human slop labels at chance.

A Claude Code skill that detects and fixes AI-generated writing patterns. 35+ patterns, tiered scoring, editor mode.

花 4 個月慢慢一步一步調整到可以自動化去防 AI Slop 的機制，包括：CLAUDE.md 跟 Hook 之間的設定跟腳本，目的就是幫助整個系統在運作的時候減少 AI Slop。

A taste system for AI agents. 288 skills, 80 agents, 12 rules, and 8 hooks that turn generic AI output into work with genuine design quality. Anti-slop detection, typography craft, color systems, brand voice, product strategy, and 50+ specialized reviewers across UI, industry, platform, and content domains.

Local-first research-agent workflows for defensible formal research outputs, with source-first gates, bounded routing, skill receipts, and delivery guards.

Claude Skills that make AI-generated prose read like a human wrote it — Persian, English, French, Italian, Spanish, Arabic, and Hebrew.

去AI味写作技能 — 5维评分体系，检测41种AI反模式。让 Claude/Cursor/Copilot 写出人话。Claude Code Skill，支持中英文。

Frankkie - Advanced roleplay prompt library with anti-slop system, psychological depth modules, and 20+ toggleable quality frameworks for AI storytelling

风格化写作工坊Python deterministic engines + ChromaDB vector retrieval = engineering-grade AI writing quality control. 5 engines detect AI-taste, validate style compliance, force argument structure, check citations, and guard logic consistency. Bilingual (ZH/EN).

Professional business writing assistant - removes AI patterns, maintains authority and clarity. For reports, proposals, emails, white papers. Works with Claude, ChatGPT, Cursor, Windsurf.

A transparent linter for AI-slop writing patterns, with a 0–100 SlopScore, evidence spans, genre-specific profiles, and a fairness audit. It identifies detectable writing patterns while avoiding claims about authorship.

Three-tier writing discipline that keeps prose from reading as AI-generated: prompt constraints, evidence-based linter, judgment checklist. Claude Code plugin + adapters for any AI client.

Anti-slop framework for Claude Code: skill + scorer + agent + hook + always-on rule. 38 AI-writing patterns, domain-aware (academic / docs / blog / commit). Extends @blader/humanizer.

Claude Code plugin that eliminates AI slop from creative output. 39 anti-patterns, 5 mechanical checks, recursive router architecture.

Privacy-first Chrome, Brave, Edge, and Firefox extension for detecting and filtering stylistic slop on LinkedIn.

Dependency-free prose-quality scanners + severity-tiered validators that detect AI-tell vocabulary, stock phrases, weak openers, poor rhythm, invented statistics, and template writing.

AI Writing Detector is an Agent Skill for reviewing AI writing signals and improving English or Brazilian Portuguese copy. It checks formulaic phrasing, punctuation, cliches, sentence rhythm, and burstiness while keeping authorship claims honest. Use it for editorial review, human-sounding edits, and safer publishing workflows. For careful editing.

Quality gate tooling for building-in-the-open artifacts

Free browser-based readability analyzer. 8 formulas, sentiment analysis, keyword extraction, SEO scoring. Privacy-first — runs entirely client-side.