[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1404":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":10,"languages":10,"totalLinesOfCode":10,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":15,"stars7d":16,"stars30d":17,"stars90d":14,"forks30d":14,"starsTrendScore":12,"compositeScore":18,"rankGlobal":10,"rankLanguage":10,"license":19,"archived":20,"fork":20,"defaultBranch":21,"hasWiki":22,"hasPages":20,"topics":23,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":14,"starSnapshotCount":14,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},1404,"WRITING.md","Anbeeld\u002FWRITING.md","Anbeeld","\"Humanizer\" rules that make AI text sharper, genre-aware, with concrete anchors and self-auditing workflow","https:\u002F\u002Fanbeeld.com\u002Fprojects\u002Fwriting-md",null,296,21,3,0,7,10,42,67.23,"MIT License",false,"main",true,[24,25,26,27,28,29],"ai-writing","content-design","llm-instructions","llm-writing","prompt-engineering","technical-writing","2026-06-12 04:00:09","# Anbeeld's WRITING.md — AI Writing Rules\n\nA prose ruleset that makes AI-assisted text sharper, not as generic, and less like default model output. It targets the patterns that make generated prose feel recognizable without breaking grammar, faking typos, or gutting useful structure. The result reads better than unguided output and avoids the most common tells.\n\nThe rules combine writing standards, web readability research, documentation guidance, detector-limit research, and direct observation of what LLMs produce when left to defaults.\n\n[![Support my work!](https:\u002F\u002Fanbeeld.com\u002Fimages\u002Fsupport.jpg)](https:\u002F\u002Fanbeeld.com\u002Fsupport)\n\n## What it does\n\n- Starts with context: medium, audience, reader need, and the job of the text.\n- Calibrates voice to genre. Opinion pieces can sound like a person; summaries and docs stay neutral.\n- Keeps the format appropriate. Docs stay scannable; casual replies do not get over-structured. Does not remove useful structure just to look less templated.\n- Replaces polished generality with concrete anchors. If a paragraph has no checkable detail, it gets revised.\n- Guards against fake specificity. Quotes, numbers, dates, causal claims, and named claims need support.\n- Breaks the patterns that make model prose look templated: repeated cadence, hidden lists, generic signposts, and too-neat paragraph shapes.\n- Prefers ordinary words over inflated phrasing. Allows repetition when the ordinary word is the right one.\n- Cuts the staged parts: keynote cadence, service-desk tone, ceremonial openings, and filler closings.\n- Does not fake humanity with some specific tricks. No invented typos, no forced slang, no programmatic sentence-length variation, no staged messiness. You can add it yourself, if needed.\n- Revises by reading and cutting. Collapses restatements, drops announcement sentences, and replaces the most generic clause with something specific.\n\n## How to use it\n\n### 1. Manually as direct instructions\n\nGive [the ruleset](WRITING.md) to an AI and tell it to follow every rule strictly. The explicit instructions (~4400 words) compensate for habits the model would otherwise default to: regular sentence structure, generic phrasing, parallel enumeration, overly neat conclusions, and many others.\n\nOne pattern that works well: ask for a first draft with the rules applied, then request an audit of that draft against the required checks, and have the model rewrite based on what the audit caught. Repeat one or two more times if the text is on a larger side. Further rewrites after that usually circulate around the same model-level limitations.\n\nThis is meant for legitimate uses, like iterating on a human draft or turning loosely organized thinking into something you would actually publish. The output avoids the most common LLM tells and will read better than unguided model text, but it will not pass an experienced eye test or an advanced detector. Prompt-level instructions have a ceiling and can't overcome model-level defaults.\n\n### 2. As a skill in an agent harness\n\nUse the packaged skill in [\u002Fskills\u002Fwriting](skills\u002Fwriting) with tools that support skills: Claude Code, Codex, OpenCode, Cursor, and others. The skill loads on demand when the agent is doing writing, so it costs almost no context until needed.\n\n**Installation:** copy the [\u002Fskills\u002Fwriting](skills\u002Fwriting) directory to your global skills folder:\n\n- **Claude Code:** `~\u002F.claude\u002Fskills\u002F`\n- **Codex:** `~\u002F.codex\u002Fskills\u002F`\n- **Cursor:** `~\u002F.cursor\u002Fskills\u002F`\n- **OpenCode:** `~\u002F.config\u002Fopencode\u002Fskills\u002F`\n\nThe skill description is scoped to public-facing prose, like blog posts, articles, documentation, criticism, long-form, emails, marketing, SEO copy, or UI text. It explicitly excludes code comments, commit messages, and private notes.\n\n### 3. Compact as agent\u002Fchat instructions\n\n[WRITING-compact.md](WRITING-compact.md) is the same rules at ~1200 words. Use as AGENTS.md or CLAUDE.md for writing agents, or as instructions in custom chats like GPTs and Gemini Gems. Cuts examples, long-form diagnostics, and provenance; keeps every behavior-changing rule, required check, and the watchlist.\n\n### 4. Mini as a section in AGENTS.md\n\n[WRITING-mini.md](WRITING-mini.md) is ~240 words, fits on one screen. All ten concepts compressed to bullet-pointed aphorisms. Designed to embed as a persistent section in global AGENTS.md or CLAUDE.md so the rules are always in context without dominating it.\n\n## See also\n\n- [Anbeeld's Global AGENTS.md \u002F CLAUDE.md](https:\u002F\u002Fgithub.com\u002FAnbeeld\u002FAGENTS.md)\n\n[![Support my work!](https:\u002F\u002Fanbeeld.com\u002Fimages\u002Fsupport.jpg)](https:\u002F\u002Fanbeeld.com\u002Fsupport)","Anbeeld的WRITING.md项目旨在通过一系列规则使AI生成的文本更加锐利、有针对性，并内置自我审计流程。其核心功能包括根据文本类型调整语气、保持格式适宜性、用具体细节替代泛泛而谈的内容、避免虚假的具体性以及打破模板化的写作模式等，从而让最终输出的文本在保持语法正确的同时更具可读性和真实性。该项目特别适合需要提高AI辅助写作质量的场景，如内容设计、技术文档撰写和技术写作等领域，能够有效提升文本的专业度和自然度。",2,"2026-06-11 02:43:33","CREATED_QUERY"]