tslj-script SKILL.md skillA
tslj-script from liuhaisi2026212-maker/taishang-laojun is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add liuhaisi2026212-maker/taishang-laojun/tslj-script@git:20261027.12e5903Pins git:20261027.12e5903 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- liuhaisi2026212-maker/taishang-laojun
- Path
skills/tslj-script/SKILL.md- Versions recorded
- 5
- Latest version
- git:20261027.12e5903, committed Oct 27, 2026
- First version recorded
- Oct 24, 2026
- Size
- 5,787 bytes
- Repository stars
- 517
- License
- a license file GitHub does not identify
回炉：完整成稿与改稿。把用户的素材或参考观点写成可以直接念的完整逐字稿，也修改已有稿件和 AI 初稿；从“我要表达什么”出发借观点，不照搬别人的脉络。用户要完整口播稿、要改稿、要把别人的爆款改成自己的时使用。.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# 回炉：完整成稿与改稿 先读[炉心](../taishang-laojun/references/core-mind.md)。 本模块负责完整逐字稿。用户要完整稿时，交可以直接念的正文，不把提纲当成稿，也不推回开炉。只要改一段就只改那一段。 ## 柳爷改稿踩过的坑 柳爷说过：自己拿别人的爆款稿去改，基本没拿到过大流量。原因是： - 抓不住重点，不知道什么该说什么不该说； - 会觉得人家数据好，是因为每句话都有道理、层层递进，删掉一句就怕破坏效果； - 结果要么改不动，要么改完和原稿脉络差不多，容易被人扒出来说模仿。 唯一一条改稿拿到大流量的是机场那条（见[案例库](../taishang-laojun/references/case-book.md)第 3 条）。它的顺序是反过来的：**先有场景和目的，再去找契合的观点**。 ## 三条经验 1. **逐字稿是弹药库**：看到感兴趣的内容，先把逐字稿存下来，不急着改。存的目的是以后找观点，不是逐句换同义词。 2. **场景优先匹配**：遇到一个好场景、一个好热点，或者特别想拍的那一刻，先问“我要通过这个场景达到什么目的”，再把收藏的逐字稿翻出来，看哪个观点最契合。已经有明确主题时，也可以从主题出发去找。 3. **借观点不借脉络**：观点相同没问题，但第一句说什么、第二句说什么、用什么例子都和人家差不多，就是照搬。先分清参考讲了什么观点、依据是什么，再用用户自己的经历、演示和判断重新组织。来源要标明，别人的原话、身份、成果不能变成用户的。 ## 最有人味的写法：先口述，再改稿 柳爷自己的爆款稿是这么来的：先有一版初稿，再由他本人一刀一刀改成自己的话。柳爷也说过，自己总结不出来，但跟人聊天能说出很多东西。所以只要用户愿意，优先用这种方式： 1. 请用户对着手机，把这件事像跟朋友聊天一样讲一遍（两三分钟就够），转成文字发来。不用讲得好，讲真的就行。 2. 你只在这份口述上动刀：删重复、调顺序、补钩子和证据，**尽量保留原话**，一句话能不改就不改。 3. 交稿时标出改动最大的两三句，请用户用自己的话再念一遍。 这样出来的稿子，主体是用户自己说的话，观众听得出是真人，也不会编出用户没有的经历。用户不想口述、只要你直接写，就按下面的流程写，交付时提醒一句：“这是初稿，你念一遍，把不像你会说的话换掉，再拍。” ## 工作流程 1. 读任务、素材、主题、目标观众和用户的说话样本。分清四种东西：原始事实、用户本人的说法、参考里的观点、需要核对的主张。 2. 有参考时，提取值得借的观点和推进逻辑，标出哪些必须换成用户自己的材料。材料缺了，问一个最关键的问题，或者明确标注占位；不编经历和数字。 …
Read the whole file at its exact version.
How to install
mdr add liuhaisi2026212-maker/taishang-laojun/tslj-script@sha256:90cbfe59bacbf687The line at the top of this page pins the label git:20261027.12e5903, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_66phbbkv52uw6ya4)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Oct 24, 2026 and Oct 27, 2026, more than one a day.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261027.12e5903 latest | 2026-10-27 | 12e5903 | 5,787 B | A | view · diff |
| git:20261026.69a46cd | 2026-10-26 | 69a46cd | 4,507 B | A | view · diff |
| git:20261026.7695ec5 | 2026-10-26 | 7695ec5 | 4,292 B | A | view · diff |
| git:20261026.8db1091 | 2026-10-26 | 8db1091 | 3,991 B | A | view · diff |
| git:20261024.0bd3156 | 2026-10-24 | 0bd3156 | 2,602 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 (5787 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
liuhaisi2026212-maker/taishang-laojun · 517 stars · license a license file GitHub does not identify · pushed 2026-10-28 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_66phbbkv52uw6ya4 GET https://markdownregistry.com/api/v1/resolve?ref=liuhaisi2026212-maker/taishang-laojun/tslj-script GET https://markdownregistry.com/api/v1/blob/90cbfe59bacbf687b11ca25780a1c4b2a05e5e9c0fa50b2fa1182783f8b3f0c5
Your AI assistant 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.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.