Home / hanzhcn / laohan-skills · laohan-chuangzuo/SKILL.md · GitHub

laohan-chuangzuo skillB

laohan-chuangzuo is agent-read markdown (skill) from hanzhcn/laohan-skills: 统一创作引擎,负责创作口播初稿(不含封面提示词,封面由 laohan-fengmianqiuzhi 独立产出;不含选题搜索,选题由工作流①前置用 laohan-xuanti/laohan-douyinsousuo 出大纲后以大纲模式喂入)。支持录屏视频(音频提取→转录)、URL队列(抓取→整理)、结构化大纲、原始文本、自由主题五种输入。其他 skill 的写作环节统一调用本 skill。Use when 用户说"写口播稿""帮我写""录屏转口播""视频转口播稿""写一篇""根据链接改写""改写文档"。.

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

# 统一创作引擎

所有涉及口播稿写作统一调用本 skill。一份方法论,一个入口。

## 核心理念

一份方法论、一个入口——所有写作走同一个管线,避免不同入口产出质量参差不齐。素材先整理(Layer A/B)再写稿,确保口播稿建立在结构化理解之上,而非对原始素材的直接改写。

内容长度由独立内容单位决定,不由预设字数或分钟数决定。内容重复与人味废话是两条不同的审计线:同一判断换词复述必须合并;结巴、语气词、自嘲、反问和停顿继续按风格保留,不能因去重被误删,也不能冒充内容增量。

## 不适用场景

- 纯文本摘要/总结 → 用 summarize skill,不走创作流程
- 翻译 → 直接翻译,不需要口播稿格式
- 学术论文/技术文档写作 → 输出格式是口播稿,不适合正式文档

## 知识来源(唯一正本,symlink 到 OpenClaw workspace)

- **写作规则**:`references/style.md`(恢复 2026-07-11 前实战版本)。其中开场、转场、修辞、标题公式、场景密度、节奏、口头禅和模板均为已验证的默认方法;允许按本题增删或扩充,但不得以“防模板化”为由整体降级成可有可无的抽象建议
- **整理方法**:`references/skill.md` → `~/.openclaw/workspace-reviewer/knowledge/skill.md`(进宝 skill.md v3.5)
- **转录方法**:`references/transcription.md`(音频提取 + 语音转文字三级降级)
- **转译选题法**:`references/yuanchuang-method.md`(转译选题法 + 角度库 + 标题公式 + 数据基准)。本 skill 不再自带热点搜索;此方法供工作流①前置阶段复用——①用 laohan-xuanti/laohan-douyinsousuo 出选题后,按本文件规则生成大纲再喂入本 skill 大纲模式
- **写作风格目录**:`references/styles/`(每份 .md 是一种写作结构框架,Step -1 强制选择)
- **创作机械合同**:`references/creation-contract.md`(Episode schema 4的选题/表达池绑定、内容单位、逐段审计、人味、自然时长、TTS与validator;独立模式兼容schema 3,历史采访合同仅对旧schema 3选题生效)
…

Read the whole file at its exact version.

How to install

Latest version
mdr add hanzhcn/laohan-skills/laohan-chuangzuo@v3.6.0
Exact content
mdr add hanzhcn/laohan-skills/laohan-chuangzuo@sha256:c40a46ff02841411

Pin 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.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_doyemjulygvrzqpt.svg)](https://markdownregistry.com/a/art_doyemjulygvrzqpt)

1 badge views in 30 days

Versions

versioncommittedcommitsizeaudit
v3.6.0 latest2026-09-18 06d5bec 39,224 BB view · diff
v3.6.02026-09-18 2923adc 39,224 BB view · diff
v3.6.02026-09-16 4b89968 36,697 BB view · diff
v3.5.02026-09-16 fa3a036 35,069 BB view · diff
v3.4.02026-08-26 de5df5b 34,962 BB view

Audit of the latest version

B  16 of 17 checks passed. Deterministic, no model, same answer every run.
  • fail: No script tag (matched: <script)
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (39224 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

Source

GitHub

hanzhcn/laohan-skills · 11 stars · license MIT · pushed 2026-09-20 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_doyemjulygvrzqpt
GET https://markdownregistry.com/api/v1/resolve?ref=hanzhcn/laohan-skills/laohan-chuangzuo
GET https://markdownregistry.com/api/v1/blob/c40a46ff02841411714895683db19d6a7922dce261ac154f376bea7a2037e535

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.

More from hanzhcn/laohan-skills

CLAUDE.md claude
hanzhcn/laohan-skills · CLAUDE.md
git:20260813.307a30f · audit A · 11 stars
duopingtai skill
hanzhcn/laohan-skills · duopingtai/SKILL.md · 将口播稿/文章改写为多平台版本。当用户说"多平台改写"、"改成多平台版本"、"改成三个平台"、"适配多平台"、"转抖音版"、"转小红书版"、"转公众号版"时触发。输入一段口播稿或文章,输出抖音、小红书、公众号三个平台的适配版本。
v1.0 · audit A · 11 stars
laohan-bianpai skill
hanzhcn/laohan-skills · laohan-bianpai/SKILL.md · 真人口播工作流编排器。根据 episode 已落盘产物识别当前步骤、验证前置 gate,并给出唯一下一步与对应 skill;不替代创作、剪辑、发布或复盘。Use when 用户说工作流下一步、检查本期进度、编排这期视频、当前做到哪、验证…
v1.23.0 · audit A · 11 stars
laohan-cheat skill
hanzhcn/laohan-skills · laohan-cheat/SKILL.md · 内容校准工作流适配器。为 episode 选择正确的 cheat-on-content 校准 lane,保存可追溯的校准入口;评分、盲预测、发布登记、复盘和 rubric 更新一律由上游 cheat-on-content 执行。Use…
v2.1.0 · audit A · 11 stars
laohan-chushicodex skill
hanzhcn/laohan-skills · laohan-chushicodex/SKILL.md · 为新安装或待整顿的 Codex 做轻量初始化:体检版本与配置、建立最小权限边界、整理分层 AGENTS.md、确认项目验证命令,并留下可重复的任务闭环。Use when 用户说"初始化 Codex"、"新装 Codex…
git:20260812.84bd329 · audit A · 11 stars
laohan-daoyan skill
hanzhcn/laohan-skills · laohan-daoyan/SKILL.md · laohanAI真人口播新episode的V5.1导演预制入口。分为导演初稿与一次最终复审,共用director-state.md并停在WAITING_FOR_FOOTAGE;不剪辑、不写Remotion、不渲染。METHOD_LAB仅在用…
v2.1.2 · audit A · 11 stars
laohan-donghua skill
hanzhcn/laohan-skills · laohan-donghua/SKILL.md · METHOD_LAB/历史 episode 的双 renderer 动画编排器。默认 CODEX_DIRECT 生产不使用本 skill;只有用户明确要求旧 AST、Remotion/HyperFrames 对照或恢复历史 episode…
v11.4.0-candidate · audit A · 11 stars
laohan-douyinsousuo skill
hanzhcn/laohan-skills · laohan-douyinsousuo/SKILL.md · 抖音关键词搜索与结构化取证 adapter,只使用已安装且已登录的 OpenCLI;独立回答搜索请求,或为真人口播①写本期抖音证据,不决定最终选题。Use when…
v3.3.0 · audit A · 11 stars
laohan-fengmianqiuzhi skill
hanzhcn/laohan-skills · laohan-fengmianqiuzhi/SKILL.md · 根据真人AI口播稿和 Jeffrey 身份参考图,按文案冲突选择1个最匹配模板,生成1张9:16视觉母版,并产出3:4/4:3/16:9三张真实尺寸封面,共享到四平台7个逻辑入口。Use when…
v1.19.0 · audit A · 11 stars
laohan-fengmianzhuzige skill
hanzhcn/laohan-skills · laohan-fengmianzhuzige/SKILL.md · 完全对标柱子哥TzFilm封面体系(81张真实封面逐张检视实证2026-09-20)——三层合成管线:AI/真实背景板(无字无人)+真人环境照片层+Remotion材质级文字层。画幅1080x1440(3:4原生)。Use when…
v6.1.0 · audit A · 11 stars
laohan-fenjingtishici skill
hanzhcn/laohan-skills · laohan-fenjingtishici/SKILL.md · 生成专业分镜图片提示词,适配扩散模型(FLUX/SDXL/Gemini)。校验质量后拆分为独立文件,或直接生成带产品描述的 Gemini 提示词模板。Use when…
v1 · audit A · 11 stars
laohan-gengxin skill
hanzhcn/laohan-skills · laohan-gengxin/SKILL.md · 工具版本检查与更新,扫描 npm/brew/pip/uv/GitHub/plugins/skills 逐项对比已装和最新版本,生成带编号报告,确认后才执行更新。Use when…
v1 · audit A · 11 stars

Every file in hanzhcn/laohan-skills

Browse by kind, by grade B, or by owner.