Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/harshaneel/humanize/ai-checknpx skills add harshaneel/humanize --skill ai-checkgit clone --depth 1 https://github.com/harshaneel/humanizeWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00081 | $0.06135 |
| Opus 5 | $0.00041 | $0.03068 |
| Sonnet 5 | $0.00016 | $0.01227 |
| Haiku 4.5 | $0.00008 | $0.00613 |
Grade A, and why
ai-check scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 506 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI-Check Skill
Forensic analysis of text for AI-generation signals. Grounded in the published detection literature (Wu et al. 2025, Mitchell et al. 2023, Kujur 2025, AAAI 2025 shared task).
The output is a structured report, not a vague judgment. Every fired signal cites evidence.
The nine signal categories
Score each category 0–3:
- 0 = No signal detected (human-consistent)
- 1 = Weak signal (possible AI, could be human)
- 2 = Moderate signal (likely AI pattern)
- 3 = Strong signal (near-certain AI pattern)
Severity-to-score mapping (use for every category):
| Evidence in category | Score |
|---|---|
| No flagged instances | 0 |
| One weak instance, or vague unease without a specific quote | 1 |
| One moderate instance, or two or more weak instances | 2 |
| One strong instance, or two or more moderate instances, or four or more weak instances | 3 |
Double-counting policy: a single phrase can fire at most two distinct signals when the phrase is genuinely diagnostic for both. Example: "it is important to note that" is both Signal A (banned vocabulary) and Signal C (institutional hedge). Log it under both, but the same phrase cannot count as two separate weak instances inside the same category.
Total score cap: 9 categories × 3 = 27 maximum.
Signal A: Perplexity (word predictability)
Look for vocabulary that is maximally safe and expected — words that are technically correct but never the most precise or interesting choice a knowledgeable human would make.
Flags:
- Generic verbs where domain-specific ones belong ("address" instead of "untangle", "implement" instead of "wire up")
- Adjectives that describe without adding information ("significant improvements", "notable progress", "key challenges")
- Hedged assertions that swap specificity for safety ("can often lead to", "may result in", "tends to")
- Any of the canonical AI vocabulary list: delve, leverage (verb), utilize, robust, comprehensive, streamline, foster, facilitate, pivotal, nuanced, notable, notably, enduring, garner, it is worth noting, it is important to note, multifaceted, in the realm of, the landscape of, a myriad of, a plethora of
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 506 lines · 81 tokens per session scan A 675276638c8f
ai-check is a skill published in the GitHub repository harshaneel/humanize (407 stars, last pushed 1mo ago), licensed MIT. It adds 81 tokens to every session and 6,135 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
humanize
AI가 쓴 한글 텍스트를 자연스럽게 윤문하는 진입 명령. humanize-korean 파이프라인을 Fast 모드(기본)로 실행하고 --strict면 정밀 3콜(진단→겨냥 윤문→finalize). 트리거 — "/humanize".
patina
Detect and rewrite AI writing patterns in Korean, English, Chinese, and Japanese text so it reads as if a human wrote it. Meaning-preservation (MPS) verified.
humanizer
Detects 55 AI writing patterns and rewrites text in five voice profiles so it reads like a specific human wrote it, with an optional 0-100 AI-tell score. Use when text sounds AI-generated or like a chatbot, when preparing a blog post, README, or LinkedIn post for publication, when auditing prose for AI tells, or when…
humanizer-ru
Проверяет русскоязычный текст на следы машинной генерации (40 regex-маркеров, у 38 полная запись доказательств) и по явной просьбе пользователя переписывает его естественным языком. Демо: vladimir-human.github.io/humanizer-ru. Отвечает на просьбы вида «очеловечь», «убери гпт-шность», «звучит как нейросеть», «проверь…
humanizer-ru
Скилл для очеловечивания русскоязычного текста. Убирает признаки AI-генерации, делает текст живым. Используй ВСЕГДА, когда пользователь просит: очеловечить текст, убрать следы нейросети, сделать текст живым/естественным, переписать как человек, humanize на русском, убрать канцелярит, убрать водянистость, сделать текст…
ai-detector
Evidence-based AI-generated text risk analysis for essays, emails, reviews, articles, messages, and other prose samples.