joeseesun/qiaomu-book-script — explained in plain English
Analysis updated 2026-05-18
Generate a Chinese spoken short video script recommending a nonfiction book, in a specific reviewer's voice.
Install the skill into an AI coding tool that supports agent skills, such as Claude Code.
Feed edited drafts back to the AI so future scripts match your personal editing style over time.
Run the local lint script to check generated scripts follow the required format.
| joeseesun/qiaomu-book-script | 16nic/comfyui-agnes-ai | 6c696e68/gpt_signup_hybrid | |
|---|---|---|---|
| Stars | 19 | 19 | 19 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | hard |
| Complexity | 2/5 | 2/5 | 4/5 |
| Audience | writer | vibe coder | developer |
Figures from each repo's GitHub metadata at analysis time.
Installs with a single command into agent skill compatible AI coding tools.
This is an agent skill for generating Chinese spoken short-video scripts that recommend nonfiction books. It is designed for AI coding tools that support agent skills, such as Claude Code. You install it with a single command and then ask your AI assistant to write a script for a book by name. The skill is built around the voice and style of a specific Chinese book reviewer called Qiaomu (向阳乔木). The output is a single complete spoken script: it opens with a branded greeting, establishes why the book's author or core argument is worth hearing, picks 3 to 5 ideas that actually change how you think about something, explains each idea using everyday work or life examples, and ends cleanly without adding marketing calls to action or extra structure labels. The skill also includes a learning loop. After the AI generates a first draft, you edit it yourself. You then give the diff back to the AI, which compares the two versions and records what you changed, so future scripts come closer to your voice with each round. A local lint script checks that the output follows the required format: one code block, correct opening, no internal labels like "point one" or estimated reading time. The README explains a common failure mode for AI book summaries: they summarize the table of contents rather than explaining why any of the ideas matter to the listener personally. This skill is explicitly designed to avoid that pattern by grounding every concept in a concrete, relatable scenario. The README is written primarily in Chinese with a short English summary section. The project is MIT licensed.
An AI agent skill that generates Chinese spoken short video scripts recommending nonfiction books in a specific reviewer's voice, and learns from your edits over time.
Mainly Python. The stack also includes Python.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly writer.
This repo across BitVibe Labs
Verify against the repo before relying on details.