podcast-writer
根据用户给出的文字或者用户指定的txt/md文件内容，从类型列表中选择合适的文本类型，按照指定的格式给出podcast直播口述稿
When & Why to Use This Skill
The podcast-writer skill is a powerful automation tool designed to transform text, documents, or web content into professional podcast scripts and audio files. It intelligently analyzes input data to select the most appropriate narrative style—ranging from news analysis to celebrity interviews—and applies specialized templates to ensure high-quality output. By refining drafts into natural, conversational language and integrating with audio generation APIs, it provides an end-to-end solution for content creators looking to streamline their podcast production workflow.
Use Cases
- Content Repurposing: Automatically convert long-form blog posts, research papers, or news articles into engaging podcast scripts for multi-channel content distribution.
- Automated News Briefings: Streamline the production of daily news podcasts by feeding the AI current event links to generate polished, spoken-word scripts and audio files.
- Educational Audio Production: Transform technical documentation or academic materials into accessible audio lessons using the 'New Technology' template for better knowledge retention.
- Interview Polishing: Process raw interview transcripts or notes into a structured 'Famous Person Interview' format, optimized for natural oral delivery and professional broadcasting.
| name | podcast-writer |
|---|---|
| description | 根据用户给出的文字或者用户指定的txt/md文件内容，从类型列表中选择合适的文本类型，按照指定的格式给出podcast直播口述稿 |
| allowed-tools | Read, Grep, Glob, Write, Search |
Your Skill Name
Instructions
1、如果用户指定了输入的txt或者md文件，那么从文件中读取输入内容{input} 2、如果用户给出一个链接，调用工具从链接中读取输入内容{input} 3、如果用户直接给出了内容，则这个内容就是输入内容{input} 4、对输入内容进行分析，判断这个内容适合使用什么样的方式来输出podcast直播口述稿，可选的类型列表如下： ["名人访谈录","中美贸易战","新闻分析","新技术","其他"] 所有无法找到明确分类的都归类为"其他" "名人访谈录"对应的提示词模版为: FAMOUS_TEMPL.md "中美贸易战"对应的提示词模版为: TRADE.md "新闻分析"对应的提示词模版为: NEWS.md "新技术"对应的提示词模版为: TECH.md "其他"对应的提示词模版为: OTHER.md 5、为这个内容生成一个合适的标题{title} 6、根据给出的提示词模版和输入{input},直接生成podcast直播口述稿，并写入当前项目目录下的/output/{title}{currentdate_currenttime}.txt 7、将{input}单独写入当前项目目录下的/output/{title}{currentdate_currenttime}original.txt 8、第6步生成的文件内容还是很拗口，请将其改成更口语化的表达，除了口述稿正文内容，不要添加其他任何标题、旁白、解释,并将结果写入当前项目目录下的/output/{title}{currentdate_currenttime}final.txt 9、生成音频文件。 运行帮助脚本runninghub_api.py,使用当前项目目录下的/output/{title}{currentdate_currenttime}final.txt 作为input.txt,使用当前项目目录下的/output/{title}{currentdate_currenttime}_final.flac 作为output.flac
# 使用指定文本文件和指定输出路径
python scripts/runninghub_api.py man input.txt output.flac