Agent skill

Sdt AI Check

by StopDisTrain in StopDisTrain/sdt-skills

检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。

MITAuto-check passedWriting & Content

Install Sdt AI Check

skills CLI
$ npx skills add StopDisTrain/sdt-skills --skill sdt-ai-check -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install StopDisTrain/sdt-skills sdt-ai-check --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/StopDisTrain/sdt-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/sdt-ai-check .claude/skills/sdt-ai-check && rm -rf skills-src

Use ~/.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/

Facts

Skill name
sdt-ai-check
GitHub stars
309
Token cost
~201 tokens
SKILL.md length
35 words
Files
3 (incl. references)
Skills in repo
15
Repo updated
First seen
Licence
MIT

At a glance

检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。

  • Works in 5 steps: 先识别体裁,读取 references/ai-signals.md… → 按文本顺序引用命中位置并标严重度。 → 另做账号语气比对:常用句长、口头词、确定程度、幽默方式、禁用表达。 → …
  • The user wants the copy to sound more human
  • SKILL.md covers 工作流, 完整生产模式的修复依据 and 单项输出
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Sdt AI Check is an agent skill from StopDisTrain/sdt-skills. 检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。 English: Check drafts for AI-writing signals and voice mismatch, including robotic phrasing, translation-like prose, fabricated vulnerability, and forced quotable lines. Use when the user wants the copy to sound more human or more like the target account.

Its SKILL.md is about 200 tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/ai-signals.md`).

It sits in Writing & Content, covering Humanizing AI text and Translation. The repository describes itself as: SDT content research, creation and quality-control skills for Codex. The licence is MIT.

When your agent uses it

  • The user wants the copy to sound more human
  • More like the target account

Example prompts

  • “/sdt-ai-check”

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. 先识别体裁,读取 references/ai-signals.md 并应用其中的误伤规则。
  2. 按文本顺序引用命中位置并标严重度。
  3. 另做账号语气比对:常用句长、口头词、确定程度、幽默方式、禁用表达。
  4. 只修复强信号和明显语气偏差;保留人的重复、犹豫和毛边。
  5. 任何修复不得新增事实、经历、结果或情绪细节。

What it can do on your machine

Read from SKILL.md and the folder at commit 0678cd0. 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

    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.

  • 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

Sdt AI Check loads about 201 tokens when it runs, and up to ~673 if it reads all its reference files. Until then it costs about 87 tokens; SKILL.md has 35 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~87
When it runs · the whole SKILL.md, loaded when a task matches
~201
With references · SKILL.md plus every file in references/, read only if the agent opens them
~673

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 StopDisTrain/sdt-skills at commit 0678cd0, republished under its MIT licence (© StopDisTrain). 35 words, ~201 tokens.

Download SKILL.mdSave it as .claude/skills/sdt-ai-check/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
sdt-ai-check
description
检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。 English: Check drafts for AI-writing signals and voice mismatch, including robotic phrasing, translation-like prose, fabricated vulnerability, and forced quotable lines. Use when the user wants the copy to sound more human or more like the target account.

SDT AI 痕迹质检

使用 SDT 自有识别清单检查机器化写作特征。默认只报告;完整 SDT 生产模式可依据账号档案和已核实原话做有限修复。

工作流

  1. 先识别体裁,读取 references/ai-signals.md 并应用其中的误伤规则。
  2. 按文本顺序引用命中位置并标严重度。
  3. 另做账号语气比对:常用句长、口头词、确定程度、幽默方式、禁用表达。
  4. 只修复强信号和明显语气偏差;保留人的重复、犹豫和毛边。
  5. 任何修复不得新增事实、经历、结果或情绪细节。

完整生产模式的修复依据

优先级:

  1. 账号本人真实口播或历史文稿
  2. sdt-account 已验证语气特征
  3. 用户本轮明确偏好
  4. 通用口语规则

没有个人语料时,只做低幅度清理,不宣称“像本人”。

单项输出

用户只要求检查时,按原文顺序引用命中位置,标记强、中、弱信号,最后总结最突出的一类问题。不要自动重写全文。

© StopDisTrain, 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 2 other files (references) in skills/sdt-ai-check of StopDisTrain/sdt-skills.

  • SKILL.md
  • agents/openai.yaml
  • references/ai-signals.md

Open the folder on GitHubat commit 0678cd0

Compare with similar skills

Sdt AI Check 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.

Sdt AI Check compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sdt AI Check this skillStopDisTrain/sdt-skills309—~201Automated safety check: PassMIT
Humanized Chinese Writing PolisherEthanYoQ/agent-xiaohongshu-workbench154—~1.1kAutomated safety check: PassMIT
Humanizer Zhai-zixun/humanizer-zh179—~1.2kAutomated safety check: PassMIT
Academic HumanizerYila-AI/awesome-research-skills133—~1.7kAutomated safety check: PassApache-2.0
Academic Paper PolishHKUSTDial/Supervisor-Skills8.8k—~3.1kAutomated safety check: PassCC-BY-NC-SA-4.0
Academic Humanizerdongshuyan/compass-skills753—~4.2kAutomated safety check: PassMIT

Similar skills

  • Humanized Chinese Writing Polisher

    EthanYoQ/agent-xiaohongshu-workbench

    A skill your agent uses when a user asks to polish, rewrite, humanize, de-AI, localize, or oralize Chinese writing, including short-video scripts, articles, social posts, speeches, newsletters, and…

    154 GitHub stars~1.1k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed
  • Humanizer Zh

    ai-zixun/humanizer-zh

    Remove signs of AI-generated, translated, or overly mechanical Chinese prose.

    179 GitHub stars~1.2k tokensUpdated 4 mo ago
    Writing & ContentAuto-check passed
  • Academic Humanizer

    Yila-AI/awesome-research-skills

    A skill your agent uses when researchers ask to remove generic, templated, or AI-like patterns from Chinese or English academic prose, make an AI-assisted draft sound more like the author's own…

    133 GitHub stars~1.7k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed
  • Academic Paper Polish

    HKUSTDial/Supervisor-Skills

    Polishes academic prose without changing its meaning: grammar and flow fixes, tone matched to the evidence, AI-tone removal and Chinese-to-English rewriting for submission.

    8.8k GitHub stars~3.1k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed
  • Academic Humanizer

    dongshuyan/compass-skills

    Draft, audit, or minimally revise English- or Chinese-language academic prose to reduce formulaic, vacuous, mechanically repetitive, or process-leaking language while preserving claims, evidence…

    753 GitHub stars~4.2k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed
  • Humanize Korean

    sangrokjung/claude-forge

    Clean up translation-ese and AI-writing tics in Korean prose so it reads like a person wrote it — content (facts, numbers, proper nouns, quotes) stays 100% unchanged, only style and rhythm are…

    852 GitHub stars~1.3k tokensUpdated 1 mo ago
    Writing & ContentAuto-check passed

More from StopDisTrain/sdt-skills

All 15 skills in this repo
  • Sdt Topic

    StopDisTrain/sdt-skills

    把已验证的传播机制转成适合目标账号的内容选题,并完成生成、迁移、评分和筛选,让选题自然连接人设、产品或门店。用户要从低粉爆款、垂直样本、邻近受众或跨赛道内容中寻找新选题时使用。

    309 GitHub stars~339 tokensUpdated 1 mo ago
    Auto-check passed
  • Sdt

    StopDisTrain/sdt-skills

    SDT 内容生产工具箱的总入口:根据当前任务选择最合适的 sdt- 模块,既能完成标题、开头等单一步骤,也能组织从账号研究到发布的完整流程。用户不知道该用哪个 SDT Skill,或希望一站式完成内容生产时使用。

    309 GitHub stars~535 tokensUpdated 1 mo ago
    Auto-check passed
  • Sdt Research

    StopDisTrain/sdt-skills

    只读检索小红书公开内容,并把账号数据、标题、封面、互动量、作者近期内容和低粉爆款整理成可追溯证据。用户要求查公开账号或笔记数据时使用;不得点赞、收藏、评论、发布、删除或导出登录凭证。

    309 GitHub stars~343 tokensUpdated 1 mo ago
    Auto-check passed
  • Sdt Wechat HTML

    StopDisTrain/sdt-skills

    把已经完成并通过质检的 Markdown 内容转换成可粘贴到微信公众号后台的 HTML,并支持选择样式、生成预览和检查粘贴兼容性。用户要制作公众号排版稿或微信发布版本时使用。

    309 GitHub stars~274 tokensUpdated 1 mo ago
    Auto-check passed
  • Sdt Xhs Title

    StopDisTrain/sdt-skills

    根据主题、正文或已有标题生成并筛选小红书标题,提供改写和 A/B 方案,同时检查账号调性、真实性以及正文能否兑现标题承诺。用户要起标题、改标题、套用标题模式或检查标题时使用。

    309 GitHub stars~247 tokensUpdated 1 mo ago
    Auto-check passed
  • Sdt Account

    StopDisTrain/sdt-skills

    把账号链接、主页截图、产品说明和历史内容整理成可复用的账号档案,明确调性、人设、受众、产品连接和表达边界。用户要求分析账号定位、建立账号资料,或开始完整内容生产流程时使用。

    309 GitHub stars~257 tokensUpdated 1 mo ago
    Auto-check passed

Questions about Sdt AI Check

What does Sdt AI Check do?

检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。. Sdt AI Check is an agent skill from StopDisTrain/sdt-skills. 检查文稿里的 AI 味和账号语气偏差,包括机器化表达、翻译腔、虚构脆弱感和过度金句。用户问文案是否像 AI 写的、是否像本人说话,或希望降低 AI 痕迹时使用。 English: Check drafts for AI-writing signals and voice mismatch, including robotic phrasing, translation-like prose, fabricated vulnerability, and forced quotable lines.

When should I use Sdt AI Check?

Sdt AI Check fits situations like: the user wants the copy to sound more human; more like the target account.

How do I install Sdt AI Check in Claude Code?

Run `npx skills add StopDisTrain/sdt-skills --skill sdt-ai-check -a claude-code`. Or copy the skill folder (skills/sdt-ai-check in StopDisTrain/sdt-skills) into .claude/skills/sdt-ai-check in your project. Claude Code loads it when a task matches its description.

How do I install Sdt AI Check in Codex?

Run `npx skills add StopDisTrain/sdt-skills --skill sdt-ai-check -a codex`. Or copy the skill folder (skills/sdt-ai-check in StopDisTrain/sdt-skills) into .agents/skills/sdt-ai-check in your project. Codex loads it when a task matches its description.

Can I use Sdt AI Check 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 StopDisTrain/sdt-skills --skill sdt-ai-check -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sdt-ai-check, .gemini/skills/sdt-ai-check, .github/skills/sdt-ai-check and .opencode/skills/sdt-ai-check in your project.

What does Sdt AI Check need to run?

SKILL.md names no scripts, command-line tools or credentials: Sdt AI Check is instructions for the agent only.

Does Sdt AI Check 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 Sdt AI Check 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 Sdt AI Check use?

Sdt AI Check 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 Sdt AI Check use?

About 201 tokens (SKILL.md is roughly 804 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 472 tokens, read only when the agent opens those files.

What are the alternatives to Sdt AI Check?

Skills that share tags, products or a category with Sdt AI Check: Humanized Chinese Writing Polisher (EthanYoQ/agent-xiaohongshu-workbench, 154 stars), Humanizer Zh (ai-zixun/humanizer-zh, 179 stars), Academic Humanizer (Yila-AI/awesome-research-skills, 133 stars) and Academic Paper Polish (HKUSTDial/Supervisor-Skills, 8.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sdt AI Check?

StopDisTrain (a GitHub user) maintains it in StopDisTrain/sdt-skills, which has 309 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on August 25, 2026.

Source: StopDisTrain/sdt-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.