AI Provenance Mark Remover
guillaumemeyer/watermarks-remover
Strips AI provenance marks from text and files: invisible Unicode, statistical text watermarks via rewriting, and C2PA, EXIF or XMP metadata across common formats.
Edit English prose for clarity, specificity, natural flow, and fit to the requested voice.
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writing --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .claude/skills && cp -r skills-src/english .claude/skills/anti-slop-writing && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .claude/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/adenaufal/anti-slop-writing/tree/main/englishType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writing --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .agents/skills && cp -r skills-src/english .agents/skills/anti-slop-writing && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .agents/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writing --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/english .cursor/skills/anti-slop-writing && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .cursor/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/adenaufal/anti-slop-writing.git --path english--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writing --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/english .gemini/skills/anti-slop-writing && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .gemini/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writingInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .github/skills && cp -r skills-src/english .github/skills/anti-slop-writing && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .github/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install adenaufal/anti-slop-writing anti-slop-writing --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/adenaufal/anti-slop-writing.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/english .opencode/skills/anti-slop-writing && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "anti-slop-writing" agent skill from https://github.com/adenaufal/anti-slop-writing/tree/main/english into .opencode/skills/anti-slop-writing/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "anti-slop-writing", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
anti-slop-writingEdit English prose for clarity, specificity, natural flow, and fit to the requested voice.
Anti Slop Writing is an agent skill from adenaufal/anti-slop-writing. Edit English prose for clarity, specificity, natural flow, and fit to the requested voice. Use for writing or revising articles, essays, posts, and other text that feels generic or padded.
Its SKILL.md is about 5.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 10 other files, including reference files (for example `AGENTS.md`, `GEMINI.md` and `SKILL-lite.md`).
It sits in Writing & Content, covering Humanizing AI text and Prompt engineering. It works with OpenAI and Google Gemini. The repository describes itself as: Stop your AI from writing like AI. A universal system prompt eliminating every known LLM style tell — works with Claude Code, Gemini CLI, Codex CLI, Copilot, Cursor, and any web… The licence is MIT.
12 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 104fae8. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Anti Slop Writing loads about 5.6k tokens when it runs, and up to ~16k if it reads all its reference files. Until then it costs about 52 tokens; SKILL.md has 3,147 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from adenaufal/anti-slop-writing at commit 104fae8, republished under its MIT licence (© adenaufal). 3,147 words, ~5,620 tokens.
.claude/skills/anti-slop-writing/SKILL.md (or your agent's skills folder). This skill also uses 9 other files; get the full folder from GitHub.Good writing serves the reader and the writer's purpose. Generic phrasing, padding, and formulaic structure can make a draft feel automated, but no single word or pattern proves who wrote it. Treat the rules below as editorial prompts: use them when they improve clarity, accuracy, or voice, and ignore them when they would distort the requested format or register.
The model names below identify recent versions, not reliable writing fingerprints. Providers update products and route requests across models; an API model ID may not match the model selected in a consumer product. Check the optional model coverage reference for dated release notes and product/API distinctions when a version-specific request requires them.
These facts describe release coverage. They do not establish distinctive prose habits for any model. Use the checks in EN-11 to edit what is actually present in a draft.
Before writing, consult references/vocabulary-banlist.md for optional vocabulary preferences and references/structural-patterns.md for editorial examples. For sentence-level repairs, use the focused language editing guide. Apply every list as a review prompt, not a detector test or a rule to alter accurate meaning.
The lists below flag wording that can become generic or padded. They are not hard bans. Keep an item when it is the accurate technical term, needed hedge, part of a quotation, requested by the user, or best fit for the format. Do not replace precise language just to avoid a word.
Significance puffers: pivotal, crucial, vital, key (as adjective), significant, essential, groundbreaking, remarkable, transformative, indelible, profound, testament, enduring, lasting, deeply rooted, paramount, indispensable, invaluable, quintessential
Analytical verbs: underscore, highlight (as "emphasize"), showcase, foster, garner, bolster, delve, embark, leverage, facilitate, utilize, encompass, cultivate, elucidate, illuminate, navigate (figurative), exemplify, embody, transcend, harness, spearhead, streamline, galvanize
Poetic nouns: tapestry, landscape (figurative), realm, paradigm, ecosystem (figurative), journey (figurative), nexus, interplay, mosaic, fabric (of society), cornerstone, beacon, pillar, catalyst, crucible, linchpin, hallmark, confluence, odyssey, trajectory, underpinning
Promotional adjectives: vibrant, rich (figurative), comprehensive, robust, seamless, innovative, dynamic, cutting-edge, meticulous, intricate, nuanced, nestled, breathtaking, renowned, diverse array, bustling, stunning, multifaceted, holistic, overarching, compelling
Puffery adverbs: seamlessly, meticulously, profoundly, intrinsically, fundamentally, remarkably, notably, crucially, undeniably, inherently, poignantly, relentlessly, tirelessly, vividly
Formal connectives to review when simpler wording fits the register: furthermore → "also" | moreover → "also" | consequently → "so" | accordingly → "so" | nonetheless → "still" | nevertheless → "still" | additionally → "also" | thus → "so" | hence → "so"
Opening/closing crutches: "In today's world," "In today's fast-paced world," "In the ever-evolving landscape of," "In an era of/where," "As we navigate the complexities of," "In conclusion," "In summary," "Overall," "It is important to note that," "It's worth noting that," "At the end of the day," "Without further ado," "In a nutshell," "The bottom line is," "Last but not least"
Copula-avoidance constructions (use "is/has" instead): serves as → is | stands as → is | marks a → describe directly | boasts → has | features (meaning "has") → has | holds the distinction of → is | emerged as → became or is | constitutes → is
Vague attribution phrases: "Experts argue," "Observers note," "Industry reports suggest," "According to some," "Many believe," "It is widely regarded," "Studies show" (without naming the study), "Research suggests" (without citing it)
Promotional phrases: "commitment to excellence," "natural beauty," "in the heart of," "rich cultural heritage," "setting the stage for," "contributing to the broader," "reflects broader trends," "paving the way for," "at the forefront of," "pushing the boundaries of," "the landscape of X is evolving," "in the realm of," "shed light on," "a game-changer"
Formulaic patterns to review when they add no needed contrast or structure: "It's not just X, it's Y" | "It's not about X, it's about Y" | "Not only X, but also Y" | "No X. No Y. Just Z." (staccato triplet) | "Whether you're [X] or [Y]..." | "From [X] to [Y], [sweeping generalization]"
Formulaic pairs to review when they add no useful distinction: "challenges and opportunities" | "on one hand... on the other hand" | "pros and cons" | "risks and rewards"
Stock conversational openers to review when they add no meaning: "But honestly?" | "Here's the truth:" | "Here's the thing:" | "Let me be clear:" | "But here's where it gets interesting..." | "Think about it this way..." | "Let me break this down..."
Common AI filler phrases (cut or replace): "plays a [crucial/key/important] role" → state the action directly | "when it comes to" → rewrite with a verb | "in order to" → "to" | "a wide range of" → name what's in the range | "needless to say" → cut | "it goes without saying" → cut | "more often than not" → "usually" or give a number | "take a closer look at" → cut or be direct | "at this point in time" → "now" | "in recent years" → give the actual years or timeframe | "dive deep into" → just discuss it | "navigate the complexities of" → name the complexities
Passive hedging constructions (cut the framing, state the fact): "It should be noted that" → cut | "It must be emphasized that" → cut | "It has been observed that" → say who observed it | "It can be argued that" → argue it or don't | "It is important to remember that" → cut | "There are several factors that" → name them
Collaborative chat artifacts (never include): "I hope this helps!" | "Of course!" | "Certainly!" | "You're absolutely right!" | "Would you like me to..." | "Is there anything else..." | "Let me know if..." | "As an AI language model..." | "I'd be happy to..." | "Great question!"
Additional phrases to review when they add no meaning: "ensuring/ensures" as padding → name the concrete action or cut | "plays a [crucial/critical/important] role in shaping" → state what it does | "highlights/supports/reflects" used vaguely → use a concrete verb or delete | "capable of X" → "able to X" or the direct verb | "rather than" where a direct comparison works → rewrite directly | "conversely" → "but" or restructure | "in essence" / "essentially" / "fundamentally" → cut when redundant | unsupported intensifiers → add evidence or delete | "one thing is clear" | "the key takeaway" | "inherent tensions" | "this raises important questions about" | stacked hedging adverbs → state the supported level of certainty
Use short, common words: "use" not "utilize," "help" not "facilitate," "show" not "demonstrate," "end" not "conclude," "start" not "embark," "dig into" not "delve into."
When you encounter a listed word that adds no meaning, don't swap it mechanically for a synonym. Restructure the sentence to say what you mean in plain language, while keeping accurate terms and claims. Empty wording can pad a sentence; a listed word by itself does not make writing weak.
Use contractions when they fit the speaker and register: "can't," "don't," "it's," "we're," "won't," "they'll," "that's." Keep full forms when they fit formal prose or the writer's voice.
Mix sentence lengths when it helps the pace and meaning. If several neighboring sentences fall into the same rhythm, revise where the repetition distracts; do not follow a quota or target word band.
Review repeated rhythms and repetitive openers. Read the passage aloud and revise only where the cadence distracts or obscures the point. Do not impose a quota or add fragments just to create variation.
A three-part list can feel formulaic when its items do not form a real group. Keep it when the content supports it; otherwise use the number of items the material calls for.
Use contrast when it clarifies a real distinction. Remove stock contrast formulas when they merely delay the point; state the claim directly.
Review "from X to Y" when it gestures at a sweeping spectrum without clarifying the claim. Keep literal ranges and established rhetorical uses when they fit the passage.
Review comma plus participial phrases such as ", highlighting the importance of...". They can add vague commentary, but can also express a real simultaneous action or result. Keep them when their subject and relationship are clear and accurate; otherwise delete, relocate, or recast the information. See language editing.
Do not append a stock “Challenges and Future Prospects” section or vague optimism about “ongoing initiatives.” End where the requested piece has made its point.
Remove “Overall,” “In conclusion,” “In summary,” or “To recap” when they only repeat what the piece already said. Use a summary when the format calls for one.
Let each paragraph hold one connected idea. Use a short paragraph for emphasis and a longer one when the reasoning needs room. Merge fragments only when they belong together, and keep lists when they serve the requested format.
Use prose when it reads more naturally. Keep headings, labeled lists, and colon-separated descriptions when the format or reader benefits from them.
Follow the user's or publication's punctuation style. If none is given, prefer periods, commas, colons, or parentheses over em and en dashes, following this project's existing style. Preserve dashes in quotations, ranges, and technical notation as required; punctuation is not evidence of authorship.
Mix declarative sentences with questions, imperatives, and deliberate fragments. Use questions, imperatives, and fragments when the purpose and register call for them. Do not add them as proof of human authorship.
Give each paragraph a clear role. A claim-evidence-implication sequence often works; use another order when it better supports the argument. Avoid repeating the same opening or conclusion without a reason.
Use simple or layered sentences according to the idea, pace, and emphasis.
Use the connector that states the relationship accurately. Do not rotate through synonyms just to create variety.
Prefer accurate, concrete terms over vague wording. Repeat a term when that is clearer than cycling through synonyms.
Review broad temporal openers such as “In today’s fast-paced world” when they delay the point. Start with the specific fact, scene, or claim when the format allows it; keep the framing when time context matters.
In informal prose, a semicolon can sound stiff. Use it when the clauses are closely related; otherwise a period or conjunction may be clearer.
Keep punctuation consistent with the requested register. Do not introduce errors or roughness to imitate a person.
Match the requested register. Shift tone only when the audience, format, or content calls for a change.
Look for sentences where the actor disappears or an introductory modifier attaches to the wrong subject. Name the actor when it matters; keep passive voice when the actor is unknown, irrelevant, or less important than the affected object. Use the language editing guide for meaning-preserving repairs.
Use passive voice when the affected object matters more than the agent. Prefer active voice when naming the agent improves clarity.
A run of punchy parallel sentences can feel like an ad slogan. Keep it when that effect is intended; otherwise state the point once.
Use contractions when they fit the requested register. Informal forms such as “gonna” or “kinda” belong only when the audience and context support them.
Use a question when the piece needs to pose a real question. Avoid rhetorical questions that only set up an obvious answer. "But does it actually work?" "Who decides that?"
Use fragments when the format and voice allow them and they add emphasis. Do not add them as a test of authorship.
Remove stacked qualifiers that add no useful precision. Keep hedging when the evidence is incomplete or the claim is uncertain.
These are editorial heuristics, not model fingerprints. They apply only when the issue is present in the draft:
For a version-specific question, check the optional model coverage reference. Release notes establish product availability and documented capabilities; they do not establish a model's writing fingerprint.
Check whether modifiers attach to the intended actor, pronouns have clear referents, clauses are joined correctly, and coordinated items have matching grammatical form. During every edit, preserve the source's actor, scope, negation, quantity, degree of certainty, and causal claim. Do not resolve ambiguity by guessing; retain it or ask for the missing context. The language editing guide gives practical repairs and conditions for retaining each form.
Replace vague claims with concrete detail when the source material supports it. Do not invent counts, places, people, dates, or examples to make prose seem specific.
Phrases such as "Experts argue," "Observers note," "According to some," and "Studies show" can hide who supports a claim. Name the source when it is known and relevant, cite the study when making a research claim, or qualify the statement. Keep an intentionally broad attribution only when the scope is accurate and useful.
Do not attach commentary that the facts do not support. Population data may or may not support a claim about community life; a founding date alone does not establish historical importance. State the evidence and keep analysis when it follows from that evidence.
Review claims that something "contributes to the broader" picture, "reflects broader trends," or has an "enduring legacy." Keep them when the relationship is explained and supported; otherwise state the evidence or remove the claim.
Avoid listing outlets as a substitute for explaining why coverage matters. Cite the relevant reporting. Mention social media activity only when it helps the reader understand the subject.
State a supported conclusion directly. Present disagreement fairly when relevant; do not manufacture balance or certainty.
State limits clearly when information is missing or uncertain. Do not add first-person uncertainty language unless it fits the speaker and context.
Do not add errors, awkward phrasing, false starts, fragments, or redundant wording to simulate a person. Keep a casual aside or self-correction only when it suits the speaker and purpose.
Keep the register suited to the audience and format. Shift it when the subject or speaker calls for a change; do not add tonal shifts to signal authenticity.
Use names, events, and examples only when they are relevant and supported by the source material or supplied context. Do not invent memories or imply personal experience.
Use first person when writing for a real speaker whose views or experiences are provided. Never invent a speaker’s experience or observation.
Use discourse markers when they fit the speaker and register:
Replace AI transitions ("Moreover," "Furthermore") with natural connectors ("And," "But," "So," "Plus," "Also," "Still," "Though").
Let the subject and speaker determine the emotional tone. For example:
When a voice guide or examples are available, follow their preferences:
Reflect the speaker’s supplied expertise and uncertainty. Do not claim expertise or insider knowledge that the context does not support.
Mix shallow and deep sentence structures. A shallow sentence: subject-verb-object, one clause. A deep sentence: multiple embeddings, subordinate clauses, parenthetical asides. Choose sentence depth for clarity, pace, and emphasis.
No writing pattern can establish who wrote a passage. AI detectors can produce false positives and false negatives, and their internal methods and performance vary by product, version, language, and text type. Do not promise that a draft will evade detection or claim that a stylistic edit proves human authorship.
Use the checklist below to improve the draft for its reader, not to game a detector.
The structural rules apply across languages, with vocabulary and idioms adapted to the target language. In this repository, the Indonesian skill is at ../indonesian/SKILL.md; packaged English-only copies should use the separate Indonesian skill rather than treating that sibling path as a dependency.
Last Updated: September 27, 2026 (v4.1) Changelog v4.1: Added actionable, meaning-preserving grammar and discourse edits; changed absolute style bans into contextual review prompts. The model coverage section remains the September 27, 2026 v4.0 snapshot.
© adenaufal, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 9 other files (references) in english of adenaufal/anti-slop-writing.
Open the folder on GitHubat commit 104fae8
Anti Slop Writing next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Anti Slop Writing this skilladenaufal/anti-slop-writing | 147 | — | ~5.6k | Automated safety check: Pass | MIT | |
| AI Provenance Mark Removerguillaumemeyer/watermarks-remover | 24k | — | ~4.9k | Automated safety check: Pass | MIT | |
| Korean AI-Text Humanizerepoko77-ai/im-not-ai | 5.9k | — | ~1.2k | Automated safety check: Pass | MIT | |
| ShuorenhuaMrGeDiao/shuorenhua | 2k | — | ~757 | Automated safety check: Pass | MIT | |
| Sloptrimseyedehsanhadi/sloptrim | 220 | — | ~5.2k | Automated safety check: Notes | Apache-2.0 | |
| WeChat Article Publisherjiji262/wechat-publisher | 274 | — | ~4.4k | Automated safety check: Pass | None |
guillaumemeyer/watermarks-remover
Strips AI provenance marks from text and files: invisible Unicode, statistical text watermarks via rewriting, and C2PA, EXIF or XMP metadata across common formats.
epoko77-ai/im-not-ai
Diagnoses and rewrites Korean text that reads as AI-generated, fixing translationese and mechanical parallelism across 85 patterns in 10 categories, with light to heavy passes.
MrGeDiao/shuorenhua
按用户要求审稿或去 AI 味,支持中英文;保留事实、术语与责任主体,支持只标问题. An agent skill from MrGeDiao/shuorenhua.
seyedehsanhadi/sloptrim
A skill your agent uses when the user wants to humanize text, trim slop, de-AI or de-slop writing, remove AI tells, fix robotic or ChatGPT-sounding prose, or make writing sound human and natural.
jiji262/wechat-publisher
Researches a topic, writes an illustrated WeChat Official Account article in a chosen author voice and sends it to the account's draft box.
cashew-labs/libretto
Lays out a minimal, iteration-first approach to writing system prompts for LLM agents, with model-specific notes for Claude, GPT, Gemini, and Codex.
adenaufal/anti-slop-writing
Tulis dan sunting teks Bahasa Indonesia agar jelas, spesifik, alami untuk audiensnya, dan sesuai suara yang diminta.
Works with
Categories
Edit English prose for clarity, specificity, natural flow, and fit to the requested voice. Anti Slop Writing is an agent skill from adenaufal/anti-slop-writing. Edit English prose for clarity, specificity, natural flow, and fit to the requested voice.
Anti Slop Writing fits situations like: revising articles; other text that feels generic.
Run `npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a claude-code`. Or copy the skill folder (english in adenaufal/anti-slop-writing) into .claude/skills/anti-slop-writing in your project. Claude Code loads it when a task matches its description.
Run `npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a codex`. Or copy the skill folder (english in adenaufal/anti-slop-writing) into .agents/skills/anti-slop-writing in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add adenaufal/anti-slop-writing --skill anti-slop-writing -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/anti-slop-writing, .gemini/skills/anti-slop-writing, .github/skills/anti-slop-writing and .opencode/skills/anti-slop-writing in your project.
SKILL.md names no scripts, command-line tools or credentials: Anti Slop Writing is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Anti Slop Writing is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
About 5.6k tokens (SKILL.md is roughly 22k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 11k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Anti Slop Writing: AI Provenance Mark Remover (guillaumemeyer/watermarks-remover, 24k stars), Korean AI-Text Humanizer (epoko77-ai/im-not-ai, 5.9k stars), Shuorenhua (MrGeDiao/shuorenhua, 2k stars) and Sloptrim (seyedehsanhadi/sloptrim, 220 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
adenaufal (a GitHub user) maintains it in adenaufal/anti-slop-writing, which has 147 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 27, 2026.
Source: adenaufal/anti-slop-writing on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.