laohan-weigui skillA
laohan-weigui is agent-read markdown (skill) from hanzhcn/laohan-skills: 抖音文案违规检测。扫描口播稿中的引流词、极限词、医疗/金融承诺、低质标记、低俗色情、引战谩骂等9类违规风险,输出结构化报告+替换建议。Use when 用户说"检测违规""检查文案""看看有没有违规词""会不会被限流""发之前帮我看看"或给出任何要发布到抖音的文案要求检查。.
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# 抖音文案违规检测 发布前扫描口播稿,找出可能触发抖音限流/禁言/下架的文案风险。它是版本化启发式筛查,不是抖音官方审核器,也不承诺发布结果。 ## 核心理念 抖音审核不是"有敏感词就违规"的简单匹配——同一个词在引用、否定、专业术语等语境下完全合法。过度标记比漏标更糟:如果每次检测都报一堆低风险,用户会养成忽略报告的习惯,真正的高风险就被淹没了。所以宁可漏标也不误标——只标记真实风险,不确定的标低风险并注明原因(高风险类别除外,见评分规则)。 ## 不适用场景 - 视频画面违规(水印、低俗画面)→ 本 skill 只管文案 - 其他平台(小红书/知乎等)→ 词库针对抖音,其他平台规则不同 - 法律合规审查 → 这是平台规则检测,不是法律审查 - 缺少文案内容 → 停下来,告诉用户需要提供文案文件或文本 ## 工作流 ### 1. 读入文案 接收文案来源(文件路径或直接文本),提取纯文本内容。 - **完成条件:** 拿到纯文本字符串 - **如果无法获取文案内容 → 停下来,告诉用户需要提供文件路径或直接粘贴文本** ### 2. 按9类扫描 先读取 `references/ruleset.json`。scan time 超过 `expires_at` 时在报告写 `ruleset_warning: STALE` 并安排复核,但不因规则过期自动阻断文案;只有命中明确未解决高风险或扫描根本无法完成时才 `REVISE_REQUIRED|BLOCKED`。再按 `references/keywords.md` 词库逐类扫描,每类独立评分。 | 类别 | 代码 | 风险等级 | 说明 | |------|------|---------|------| | 引流违规 | YL | 高 | 引导用户加微信/扫码/私信等站外引流(含谐音变形:薇信/徵信/加V♥等) | | 极限词/广告法 | JX | 高 | 最、第一、唯一等绝对化用语 | | 医疗效果承诺 | YL-MED | 高 | 治愈、根治、100%有效等医疗声明 | | 金融收益承诺 | YL-FIN | 高 | 保收益、稳赚、具体收益数字 | | 低质/标题党 | DZ | 中 | 震惊体、悬念杀、信息空洞 | | 敏感话题 | MG | 高 | 政治敏感、未经证实消息、争议性话题 | | 平台特有限制 | PT | 中 | 抖音独有的隐性限流规则(重复发布、诱导互动等) | | 低俗色情 | DS | 高 | 抖音第一红线:擦边/性暗示/软色情即限流,硬色情封号 | | 引战谩骂 | ZH | 高 | 人身攻击/地域攻击/拉踩引战/脏话,AI测评易踩拉踩线 | - **完成条件:** 9类全部扫描完毕,记录所有匹配项及其上下文 ### 3. 输出报告 …
Read the whole file at its exact version.
How to install
mdr add hanzhcn/laohan-skills/laohan-weigui@v1.3mdr add hanzhcn/laohan-skills/laohan-weigui@sha256:5f047e645c670883Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_hjowcqabn6tlmjjm)
1 badge views in 30 days
Versions
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.3 latest | 2026-07-20 | eeb59f4 | 7,859 B | A | view · diff |
| v1.3 | 2026-07-20 | 446cb18 | 7,814 B | A | view · diff |
| v1.2 | 2026-07-11 | a268666 | 7,757 B | A | view · diff |
| v1.1 | 2026-07-10 | a229557 | 6,751 B | A | view · diff |
| v1 | 2026-05-22 | 41064ee | 5,177 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (7859 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
hanzhcn/laohan-skills · 11 stars · license MIT · pushed 2026-09-20 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_hjowcqabn6tlmjjm GET https://markdownregistry.com/api/v1/resolve?ref=hanzhcn/laohan-skills/laohan-weigui GET https://markdownregistry.com/api/v1/blob/5f047e645c670883cc1ff016d54a66c6d5eee39198d4f388fd49f00328ec5b4e
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.