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 skills add Desko77/claude-code-skills-1c --skill humanize-ai-textgit clone --depth 1 https://github.com/Desko77/claude-code-skills-1cWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/desko77/claude-code-skills-1c/humanize-ai-text)<a href="https://agentmods.dev/skills/desko77/claude-code-skills-1c/humanize-ai-text"><img src="https://agentmods.dev/badge/skills/desko77/claude-code-skills-1c/humanize-ai-text/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/desko77/claude-code-skills-1c/humanize-ai-text"><img src="https://agentmods.dev/badge/skills/desko77/claude-code-skills-1c/humanize-ai-text.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00329 | $0.05872 |
| Opus 5 | $0.00164 | $0.02936 |
| Sonnet 5 | $0.00066 | $0.01174 |
| Haiku 4.5 | $0.00033 | $0.00587 |
Grade A, and why
humanize-ai-text 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 5d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- humanize-ai-text — 86% identical, 40 lines differ
How it starts
The opening of the file, as written. The whole thing — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/humanize-ai-text - переписывание AI-текста в живой стиль
Превращает текст с маркерами LLM-генерации (ровный ритм, лестница H2/H3, буллеты вместо прозы, дежурные вступления и заключения) в текст с человеческой интонацией, не теряя смысл, числа и термины.
Когда использовать
Триггерные фразы пользователя:
- "убери AI-стиль", "не как ChatGPT", "не как нейросеть"
- "перепиши по-человечески", "сделай естественно", "сделай живым"
- "убери воду", "убери штампы", "убери LLM-маркеры"
- "причеши текст", "оживи текст"
Автотриггер при анализе входного текста:
- Заголовки H2/H3 на каждый второй абзац в коротком документе.
- Буллеты с симметричной структурой и одинаковой длиной пунктов.
- Стоп-фразы из таблицы ниже ("в современном мире", "давайте погрузимся" и т.п.).
- Эмодзи-маркеры в начале пунктов (галочки, ракеты, стрелки).
- Дежурные "надеюсь, это поможет!" / "дайте знать, если есть вопросы!".
- Серия предложений одинаковой длины подряд (4+).
Режимы работы
| Режим | Триггер | Что делает |
|---|---|---|
| Inline | аргумент - произвольный текст | Переписанный текст выводится в чат |
| File | аргумент - путь к существующему .md файлу | Результат сохраняется рядом с суффиксом -human.md |
| Interactive | аргумент пустой | Спросить у пользователя текст или путь |
| Встроенный | скил вызван другим агентом как шаг задачи | Отдать ТОЛЬКО итоговый текст |
Встроенный режим - когда результат идет дальше в чужую работу: описание pull request, сообщение коммита, кусок документации. Ни черновика, ни разбора, ни резюме правок: вызывающему нужен текст, а не отчет о переписывании.
Алгоритм определения режима - как в /prompt-enhancer:
- Пустой аргумент - Interactive.
- Read удалось прочитать аргумент - File.
- Read вернул "не найден" - Inline (аргумент целиком как текст).
Технический регистр: включен по умолчанию
Сверх поиска признаков генерации скил проверяет ТЕХНИЧЕСКИЙ РЕГИСТР. Проверка включена всегда,
отключается ключом --no-technical либо словами пользователя "без проверки регистра".
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 5d ago First seen · 315 lines · 329 tokens per session scan A e1a95bf5dfd8
humanize-ai-text is a skill published in the GitHub repository Desko77/claude-code-skills-1c (60 stars, last pushed 7d ago), licensed MIT. It adds 329 tokens to every session and 5,872 once invoked, about $0.0016 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 2026-09-03.
Other skills, from other repositories
docx-from-sample
A document-building tool that creates a new DOCX file using an existing DOCX as its formatting template. DOCX is the file format used by Microsoft Word.
1c-bsp-registration
A helper for 1C external reports and processors that adds the function needed to register them in the Standard Subsystems Library (БСП). БСП is a set of reusable features for 1C configurations.
claude-md-bootstrap
A project setup tool that creates or updates a CLAUDE.md file, which gives an AI coding agent project-specific instructions and background at the start of each session.
1c-cfe-patch-method
A tool for creating method interceptors in a 1C configuration extension. 1C is business software, and an interceptor can run code before, after, or instead of an inherited method.
v8unpack-cf
A Python tool for unpacking and rebuilding 1C configuration files (CF, CFE, and EPF) as readable JSON and BSL source files. 1C is a business software platform widely used in Russian-speaking companies.
1c-meta-compile
A generator that creates 1C metadata objects from a JSON description. 1C is a platform for building business applications, and metadata defines objects such as directories, documents, registers, reports, and modules.