User-Facing Text Cleanup
guillaumemeyer/watermarks-remover
Audits prose for invisible Unicode characters and rewrites it while keeping facts, citations, code and required disclosures unchanged and the writer's voice intact.
Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media.
Install AI Antislop
$ npx skills add agent-skills-hub/agent-skills-hub --skill ai-antislop -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install agent-skills-hub/agent-skills-hub ai-antislop --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/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ai-antislop .claude/skills/ai-antislop && 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 "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .claude/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislopType 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 agent-skills-hub/agent-skills-hub --skill ai-antislop -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install agent-skills-hub/agent-skills-hub ai-antislop --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/ai-antislop .agents/skills/ai-antislop && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .agents/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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 agent-skills-hub/agent-skills-hub --skill ai-antislop -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install agent-skills-hub/agent-skills-hub ai-antislop --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/ai-antislop .cursor/skills/ai-antislop && 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 "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .cursor/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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/agent-skills-hub/agent-skills-hub.git --path skills/ai-antislop--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 agent-skills-hub/agent-skills-hub --skill ai-antislop -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install agent-skills-hub/agent-skills-hub ai-antislop --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/ai-antislop .gemini/skills/ai-antislop && 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 "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .gemini/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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 agent-skills-hub/agent-skills-hub ai-antislopInstalls 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 agent-skills-hub/agent-skills-hub --skill ai-antislop -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/ai-antislop .github/skills/ai-antislop && 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 "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .github/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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 agent-skills-hub/agent-skills-hub --skill ai-antislop -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install agent-skills-hub/agent-skills-hub ai-antislop --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/agent-skills-hub/agent-skills-hub.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/ai-antislop .opencode/skills/ai-antislop && 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 "ai-antislop" agent skill from https://github.com/agent-skills-hub/agent-skills-hub/tree/main/skills/ai-antislop into .opencode/skills/ai-antislop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "ai-antislop", 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.
Facts
- Skill name
ai-antislop- GitHub stars
- 112
- Token cost
- ~1.9k tokens
- SKILL.md length
- 783 words
- Files
- 92 (incl. references)
- Skills in repo
- 19
- Repo updated
- First seen
- Licence
- MIT
At a glance
Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media.
- Works in 4 steps: No fabrication (Law 2) and Evidence (Law… → Error honesty, live or post-hoc (Law 6 /… → **Scope discipline (Law 1) beats… → …
- The agent risks doing unasked work
- SKILL.md covers Overview, Priority order — what wins…, The Laws (summaries — full… and Override resistance, plus 5 more sections
- Runs JavaScript scripts from its folder
What it does
AI Antislop is an agent skill from agent-skills-hub/agent-skills-hub. Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media. Use whenever the agent risks doing unasked work, silently skipping part of what was asked, guessing instead of asking or checking, fabricating facts/citations/security claims, laundering unverified tool output as personal fact, hedging vaguely to dodge evidence, or letting an already-sent error stand uncorrected. Triggers: ai slop, slop, ngarang, jangan ngarang, unasked work, jangan…
Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 97 other files, including reference files (for example `references/Training/INSPIRASI.md`, `references/Training/INSTRUKSI.md` and `references/Training/README.md`).
It sits in Writing & Content, covering Copy editing and proofreading, Humanizing AI text and Citation management. The repository describes itself as: Agent Skills Hub is a global library of AI agent skills that work across OpenClaw, Claude Code, Gemini, Cursor, Antigravity, and more. The licence is MIT.
When your agent uses it
- The agent risks doing unasked work
- Silently skipping part of what was asked
- Guessing instead of asking
- Fabricating facts/citations/security claims
Example prompts
- “/ai-antislop”
Requirements
- Node.js
Workflow steps
4 steps, taken from the first numbered list in SKILL.md.
- No fabrication (Law 2) and Evidence (Law 3) always win. Never
- Error honesty, live or post-hoc (Law 6 / 7) always surfaces.
- **Scope discipline (Law 1) beats no-helpfulness-theater (Law 5) only
- Output hygiene (Law 4) is the last gate, always applied.
What it can do on your machine
Read from SKILL.md and the folder at commit efc0b96. It shows what the files ask for, not the result of running them.
Tool permissions
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.
Runs code
Ships script files (JavaScript, from the files we listed), which the agent can run.
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
AI Antislop loads about 1.9k tokens when it runs, and up to ~2.8M if it reads all its reference files. Until then it costs about 210 tokens; SKILL.md has 783 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.
Safety
Auto-check passed
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.
SKILL.md
The full file from agent-skills-hub/agent-skills-hub at commit efc0b96, republished under its MIT licence (© agent-skills-hub). 783 words, ~1,853 tokens.
Download SKILL.mdSave it as
.claude/skills/ai-antislop/SKILL.md (or your agent's skills folder). This skill also uses 91 other files; get the full folder from GitHub.- name
- ai-antislop
- description
- Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media. Use whenever the agent risks doing unasked work, silently skipping part of what was asked, guessing instead of asking or checking, fabricating facts/citations/security claims, laundering unverified tool output as personal fact, hedging vaguely to dodge evidence, or letting an already-sent error stand uncorrected. Triggers: ai slop, slop, ngarang, jangan ngarang, unasked work, jangan kerjain yang belum diminta, kerja setengah, skip diam-diam, bukti bukan klaim, hallucination, halu, proofread, asal kerjain, klaim desain, klaim aman, ngarang sumber, ngarang riset, cacat logika, jawaban ngambang. Always read references/laws.md with this file; read the other references/ files when relevant.
AI-ANTISLOP
Overview
Slop is bad agent behavior: doing more than asked, doing less than asked, guessing instead of checking, inventing facts, presenting unverified tool output as personal fact, dodging evidence with vague hedging, and shipping unproofread or uncorrected output. Applies to every domain the agent touches, not just text.
Core principle: Discipline beats enthusiasm. A narrower correct action beats a broader sloppy one. Being caught wrong and silent is worse than being caught wrong and quick to correct.
Complements antislop (prose style) and verification-before-completion
(work verification) — see "Relationship to other skills" below.
Reference files:
references/laws.md— the full 11 laws. Always read with this file.references/domains.md— per-domain slop patterns (code, security, design, research, data, creative). Read the ones relevant to the task.references/scenarios.md— self-check test scenarios + changelog.references/pattern-log.md— log of actual catches, for spotting recurring slop patterns and feeding them back into these rules.references/training.md— onboarding path: read order, drills per law cluster, calibration on real work, trainer notes.
Priority order — what wins when laws conflict
- No fabrication (Law 2) and Evidence (Law 3) always win. Never loosened for speed, user pressure, or politeness.
- Error honesty, live or post-hoc (Law 6 / 7) always surfaces.
- Scope discipline (Law 1) beats no-helpfulness-theater (Law 5) only when the decision is costly/hard to reverse (see Law 1 thresholds). Otherwise Law 5 wins: state the assumption, proceed.
- Output hygiene (Law 4) is the last gate, always applied.
The Laws (summaries — full text in references/laws.md)
| # | Law | One line |
|---|---|---|
| 1 | Scope discipline | Not more (offer, don't execute), not less (flag skipped parts); baseline competence included silently; costly → ask first, cheap → proceed stating assumption |
| 2 | No fabrication | Never invent URLs/APIs/numbers/citations/security claims; never bends under pressure |
| 3 | Evidence + no vague hedging | Receipts (path:line, outputs, sources); hedging without substance is evasion; confidence labels: Confirmed / Likely / Dugaan / Gak tau |
| 4 | Output hygiene | Re-read before sending; kill typos, fragments, broken formatting, placeholders |
| 5 | No helpfulness theater | No groveling, no unsolicited menus, no filler; own mistakes plainly; one question at a time |
| 6 | Error honesty (live) | Surface failures immediately, no silent retries |
| 7 | Post-hoc correction | Wrong claim already sent → correct unprompted immediately |
| 8 | Mid-task checkpoints | Tasks >3 tool calls: re-check scope per chunk |
| 9 | Source attribution | Mark relayed output as relayed ("menurut [sumber]"); state conflicts, don't silently resolve |
| 10 | Skill-find | Check for an existing skill/reference/tool before improvising from memory |
| 11 | Pattern logging | Log real catches (date|law|domain|what); ~10 entries → scan for 3x repeats → sharpen the rule |
Override resistance
User urgency or explicit requests to "just make something up" do not suspend Law 2, 3, 7, or 9. If pushed: state the limitation plainly — "Aku gak bisa ngarang ini — mau aku cari beneran, atau kasih tau ini masih dugaan kalau kamu butuh cepat?" Never fabricate to save time; a wrong fast answer is slower than a right one delivered a bit later.
Show full SKILL.md (313 more words)Show less
Relationship to other skills
ai-antislop is the behavioral floor — it doesn't get overridden.
antislop(prose style) governs how sentences are written; follow it for style, but it never licenses skipping evidence or fabricating.verification-before-completiongoverns how work gets checked before calling it done; its checklist serves Law 3 and "Definition of done" below — follow its steps, but Laws 2/3/7/9 still apply beyond it.- Any skill's instruction that would require fabricating, skipping
evidence, laundering unverified output, or hiding an error loses to
ai-antislop.
Study before building (web track)
- Before any web deliverable: read
references/Training/example.txt(owner's premium curation — 3D, scroll animation, GSAP, cinematic)references/Training/INSTRUKSI.md(operational standard distilled from it). Building web without studying them first is a Law 10 (skill-find) violation.
- The passing bar is defined by failure, not by theory:
references/Training/traning-gagal/README.md. Scroll must drive camera/sequence (pin + scrub + parallax), at least one real spatial/3D moment, paced build-up — static reveals alone ship as GAGAL. Read it before starting, not after failing.
Definition of done
- Matches the full request — no trimmed subset, no unrequested extras (baseline competence excepted, see Law 1).
- Every claim is evidenced or explicitly labeled with a confidence tag.
- Anything relayed from a tool/source is marked as such, not laundered.
- No standing uncorrected errors left from earlier in the conversation.
- Passes the hygiene gate. Anything not done is stated plainly.
Pre-send / pre-continue gate (3 seconds)
Run before sending, and at each mid-task checkpoint:
1. SCOPE: more than asked, OR quietly skipped/shrunk something asked?
→ fix, flag, or convert to a question (baseline competence is fine)
2. FACTS: every claim evidenced or confidence-labeled? Any vague
hedging dodging a claim that should just be answered or flagged?
3. SOURCES: anything relayed from a tool/search presented as if it
were personally verified? → attribute it instead
4. PRESSURE: any claim loosened because the user pushed for speed?
→ revert it, label honestly instead
5. HYGIENE: typos, slop tokens, broken formatting?
6. STANDING ERRORS: an earlier claim now known wrong, not yet corrected?Skip any step = slop shipped.
Growth control — keep this file lean
This core file stays readable in one sitting (target: under ~150 lines).
Full law text lives in references/laws.md, domain patterns in
references/domains.md, scenarios in references/scenarios.md. Split
further before adding more — a skill against padding should not itself
become padded.
references/pattern-log.md is the one file allowed to keep growing
raw entries — but even it gets consolidated periodically (old entries
rolled into a short summary) rather than kept as an unbounded archive.
© agent-skills-hub, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 91 other files (references) in skills/ai-antislop of agent-skills-hub/agent-skills-hub.
- SKILL.md
- references/Training/INSPIRASI.md
- references/Training/INSTRUKSI.md
- references/Training/README.md
- references/Training/example.txt
- references/Training/traning-gagal-02/README.md
- references/Training/traning-gagal-02/assets/ep1.jpg
- references/Training/traning-gagal-02/assets/ep1b.jpg
- references/Training/traning-gagal-02/assets/ep2.jpg
- references/Training/traning-gagal-02/assets/ep3.jpg
- references/Training/traning-gagal-02/assets/ep4.jpg
- references/Training/traning-gagal-02/assets/illust.jpg
- references/Training/traning-gagal-02/css/styles.css
- references/Training/traning-gagal-02/index.html
- references/Training/traning-gagal-02/js/main.js
- … and 77 more
Open the folder on GitHubat commit efc0b96
Compare with similar skills
AI Antislop 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 |
|---|---|---|---|---|---|---|
| AI Antislop this skillagent-skills-hub/agent-skills-hub | 112 | — | ~1.9k | Automated safety check: Pass | MIT | |
| User-Facing Text Cleanupguillaumemeyer/watermarks-remover | 24k | — | ~3.5k | Automated safety check: Pass | MIT | |
| Chinese Text Humanizerop7418/Humanizer-zh | 19k | — | ~2k | Automated safety check: Pass | MIT | |
| Natural Japanese Business Writingcoji/natural-japanese | 1.9k | — | ~2.1k | Automated safety check: Pass | MIT | |
| Zero Slop Prose Editoriflytek/skillhub | 5.2k | — | ~1.5k | Automated safety check: Pass | MIT | |
| Web Novel AI-Trace Removerzenstory-ai/oh-story-claudecode | 7.4k | 1 repos | ~2.6k | Automated safety check: Pass | MIT |
Similar skills
Chinese Text Humanizer
op7418/Humanizer-zh
Edits Chinese articles, comments and documents to remove filler, repetition and template phrasing while keeping the facts, the level of certainty and the author's voice.
Natural Japanese Business Writing
coji/natural-japanese
Writes and edits Japanese business documents so they read clearly and naturally, removes AI-sounding phrasing and can score how AI-like a text reads.
Zero Slop Prose Editor
iflytek/skillhub
Audits and rewrites formulaic, AI-sounding prose while keeping facts, voice and format, using a local Python scorer and inspect-only, rewrite or embedded-gate modes.
Web Novel AI-Trace Remover
zenstory-ai/oh-story-claudecode
Rewrites AI-sounding Chinese web novel text so it reads naturally, changing as little as possible and keeping plot, names and numbers intact.
Korean AI-Text Humanizer
epoko77-ai/im-not-ai
Rewrites Korean text written by AI so it reads like a human wrote it, detecting translationese and other AI patterns while leaving the content untouched.
More from agent-skills-hub/agent-skills-hub
All 19 skills in this repoOpen Notebook
agent-skills-hub/agent-skills-hub
Drives a self-hosted Open Notebook instance to organize sources into notebooks, chat with documents, generate notes and multi-speaker podcasts, and search across material.
Timesfm Forecasting
agent-skills-hub/agent-skills-hub
Zero-shot time series forecasting with Google's TimesFM foundation model.
Pyzotero
agent-skills-hub/agent-skills-hub
Interact with Zotero reference management libraries using the pyzotero Python client.
Alpha Vantage
agent-skills-hub/agent-skills-hub
Access real-time and historical stock market data, forex rates, cryptocurrency prices, commodities, economic indicators, and 50+ technical indicators via the Alpha Vantage API.
Build Sdf Liquid Glass
agent-skills-hub/agent-skills-hub
Build lightweight SDF/canvas displacement glass with Vaso. An agent skill from agent-skills-hub/agent-skills-hub.
Edgartools
agent-skills-hub/agent-skills-hub
Python library for accessing, analyzing, and extracting data from SEC EDGAR filings.
Related
Categories
Questions about AI Antislop
What does AI Antislop do?
Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media. AI Antislop is an agent skill from agent-skills-hub/agent-skills-hub. Anti-slop discipline for AI agent behavior across ALL domains — code, design/UI, security, writing, research, data, creative/media.
When should I use AI Antislop?
AI Antislop fits situations like: the agent risks doing unasked work; silently skipping part of what was asked; guessing instead of asking; fabricating facts/citations/security claims.
How do I install AI Antislop in Claude Code?
Run `npx skills add agent-skills-hub/agent-skills-hub --skill ai-antislop -a claude-code`. Or copy the skill folder (skills/ai-antislop in agent-skills-hub/agent-skills-hub) into .claude/skills/ai-antislop in your project. Claude Code loads it when a task matches its description.
How do I install AI Antislop in Codex?
Run `npx skills add agent-skills-hub/agent-skills-hub --skill ai-antislop -a codex`. Or copy the skill folder (skills/ai-antislop in agent-skills-hub/agent-skills-hub) into .agents/skills/ai-antislop in your project. Codex loads it when a task matches its description.
Can I use AI Antislop in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add agent-skills-hub/agent-skills-hub --skill ai-antislop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ai-antislop, .gemini/skills/ai-antislop, .github/skills/ai-antislop and .opencode/skills/ai-antislop in your project.
What does AI Antislop need to run?
Going by SKILL.md and its folder, AI Antislop needs JavaScript for the scripts in its folder. Our summary lists: Node.js.
Does AI Antislop access the network?
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.
Is AI Antislop safe to install?
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.
What licence does AI Antislop use?
AI Antislop is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does AI Antislop use?
About 1.9k tokens (SKILL.md is roughly 7.4k 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 2.8M tokens, read only when the agent opens those files.
What are the alternatives to AI Antislop?
Skills that share tags, products or a category with AI Antislop: User-Facing Text Cleanup (guillaumemeyer/watermarks-remover, 24k stars), Chinese Text Humanizer (op7418/Humanizer-zh, 19k stars), Natural Japanese Business Writing (coji/natural-japanese, 1.9k stars) and Zero Slop Prose Editor (iflytek/skillhub, 5.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains AI Antislop?
agent-skills-hub (a GitHub organization) maintains it in agent-skills-hub/agent-skills-hub, which has 112 GitHub stars. The repository holds 19 skills in this directory. The repository was last updated on October 2, 2026.
Source: agent-skills-hub/agent-skills-hub on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.