daily-news-caster

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Daily News Caster Skill

每日新闻播报技能

This skill allows the agent to fetch real-time news, organize it into a conversational podcast script, and generate an audio file reading the script out loud.

本技能可让Agent获取实时新闻，将其整理为对话式播客脚本，并生成朗读该脚本的音频文件。

Workflow Instructions

工作流程说明

When the user asks to get the latest news and make a podcast out of it, follow these steps strictly:

当用户要求获取最新新闻并将其制作成播客时，请严格遵循以下步骤：

Step 1: Ensure Skills are Installed

步骤1：确认技能已安装

If the

news-aggregator-skill

and

tts

skills are not already installed in the workspace, run the following commands to install them:

bash

npx skills add https://github.com/cclank/news-aggregator-skill --skill news-aggregator-skill -y
npx skills add https://github.com/noizai/skills --skill tts -y

如果工作区中尚未安装

news-aggregator-skill

和

tts

技能，请运行以下命令进行安装：

bash

npx skills add https://github.com/cclank/news-aggregator-skill --skill news-aggregator-skill -y
npx skills add https://github.com/noizai/skills --skill tts -y

Step 2: Fetch the Latest News

步骤2：获取最新新闻

Find the

fetch_news.py

script from the

news-aggregator-skill

(usually located in

.cursor/skills/news-aggregator-skill/scripts/fetch_news.py

skills/news-aggregator-skill/scripts/fetch_news.py

Run the script to fetch real-time news. You can specify a source (e.g.,

hackernews

github

all

) or keywords based on the user's request. Example command:

bash

python3 path/to/fetch_news.py --source all --limit 10 --deep

在

news-aggregator-skill

中找到

fetch_news.py

脚本（通常位于

.cursor/skills/news-aggregator-skill/scripts/fetch_news.py

或

skills/news-aggregator-skill/scripts/fetch_news.py

路径下）。

运行该脚本以获取实时新闻。你可以根据用户需求指定新闻来源（例如

hackernews

、

github

、

all

）或关键词。示例命令：

bash

python3 path/to/fetch_news.py --source all --limit 10 --deep

Step 3: Write the Podcast Script

步骤3：撰写播客脚本

Read the fetched news data and rewrite the information into a Markdown podcast script. Crucially, prioritize a dual-host (two-person) conversational format (e.g., Host A and Host B) in a dynamic Q&A style. The script should be:

Dual-Host Conversational yet concise: Write an engaging back-and-forth between two hosts. Host A should ask insightful, high-value questions to guide the conversation, and Host B should provide informative, concise answers. It should feel like a smart, fast-paced Q&A dialogue.
Avoid fluff: Do not include unnecessary fluff or overly long transitions. Keep it to the point (言简意赅) while retaining all critical information and facts.
Clearly Labeled Speakers: Start each line or paragraph with the speaker's name (e.g.,
```
Host A:
```
or
```
Host B:
```
).
Structured: Use markdown headings and clear paragraphs.
Clear text for speech: Avoid complex URLs, raw markdown links, or unpronounceable characters in the spoken text.

Save this script to a local file named

podcast_script.md

Example
podcast_script.md
Content:

markdown

**Host A:** Welcome to today's news roundup. We have some exciting tech updates today. To start things off, there's a big update from [Company Name]. What are the core implications of their new release for everyday users?

**Host B:** The main takeaway is that... [Insert concise answer and summary of News Item 1]. This completely changes how we approach [Topic].

**Host A:** That's fascinating. But does this new approach raise any security concerns, especially given recent data breaches?

**Host B:** Exactly. Experts are pointing out that... [Insert analysis or context]. 

**Host A:** Moving on to the open-source world, what's trending on GitHub today that developers should pay attention to?

**Host B:** A standout project is... [Insert concise summary of News Item 2].

**Host A:** Great insights. That's all for today's quick update. Thanks for tuning in!

读取获取到的新闻数据，将信息重写为Markdown格式的播客脚本。核心要求：优先采用双主持人（双人）对话格式（例如Host A和Host B），使用生动的问答风格。脚本需要满足以下要求：

双主持人对话且简洁： 撰写两个主持人之间有吸引力的互动内容。Host A需提出有洞察力、高价值的问题引导对话，Host B需提供信息丰富、简洁的回答。整体要呈现出睿智、快节奏的问答对话感。
避免冗余内容： 不要加入无意义的废话或过长的过渡内容。在保留所有关键信息和事实的前提下做到言简意赅。
清晰标注发言者： 每一行或段落开头标注发言者姓名（例如
```
Host A:
```
或
```
Host B:
```
）。
结构清晰： 使用Markdown标题和清晰的段落划分。
适合朗读的文本： 朗读文本中避免出现复杂URL、原生Markdown链接或无法发音的字符。

将该脚本保存为本地文件，命名为

podcast_script.md

。

podcast_script.md
内容示例：

markdown

**Host A:** Welcome to today's news roundup. We have some exciting tech updates today. To start things off, there's a big update from [Company Name]. What are the core implications of their new release for everyday users?

**Host B:** The main takeaway is that... [Insert concise answer and summary of News Item 1]. This completely changes how we approach [Topic].

**Host A:** That's fascinating. But does this new approach raise any security concerns, especially given recent data breaches?

**Host B:** Exactly. Experts are pointing out that... [Insert analysis or context]. 

**Host A:** Moving on to the open-source world, what's trending on GitHub today that developers should pay attention to?

**Host B:** A standout project is... [Insert concise summary of News Item 2].

**Host A:** Great insights. That's all for today's quick update. Thanks for tuning in!

Step 4: Generate the Podcast Audio

步骤4：生成播客音频

Use the

tts

skill to convert the script into an audio file. Since it's a dual-host format, you must use the Timeline Mode of the

tts

skill.

Find the

tts.sh

script (usually located in

skills/tts/scripts/tts.sh

.cursor/skills/tts/scripts/tts.sh

1. Generate SRT: Convert the script to SRT format.

bash

bash path/to/tts.sh to-srt -i podcast_script.md -o podcast.srt

2. Create a Voice Map (
voice_map.json
): Map the lines in the SRT to the corresponding host. If the user provided reference audio files or URLs for the two roles, you MUST use them via the

reference_audio

field (requires

noiz

backend). Otherwise, assign default distinct voices.

json

{
  "default": { "voice_id": "voice_host_A", "target_lang": "zh" },
  "segments": {
    "1": { "reference_audio": "path/or/url_to_user_audio_for_host_A.wav" },
    "2": { "reference_audio": "path/or/url_to_user_audio_for_host_B.wav" },
    "3": { "reference_audio": "path/or/url_to_user_audio_for_host_A.wav" }
  }
}

(Adjust the segment numbers corresponding to the generated SRT lines for Host A and Host B).

3. Render Audio:

bash

undefined

使用

tts

技能将脚本转换为音频文件。由于是双主持人格式，你必须使用

tts

技能的时间线模式。

找到

tts.sh

脚本（通常位于

skills/tts/scripts/tts.sh

或

.cursor/skills/tts/scripts/tts.sh

路径下）。

1. 生成SRT: 将脚本转换为SRT格式。

bash

bash path/to/tts.sh to-srt -i podcast_script.md -o podcast.srt

2. 创建音色映射（
voice_map.json
）: 将SRT中的台词映射到对应的主持人。如果用户为两个角色提供了参考音频文件或URL，你必须通过

reference_audio

字段使用它们（需要

noiz

后端支持）。否则分配默认的不同音色。

json

{
  "default": { "voice_id": "voice_host_A", "target_lang": "zh" },
  "segments": {
    "1": { "reference_audio": "path/or/url_to_user_audio_for_host_A.wav" },
    "2": { "reference_audio": "path/or/url_to_user_audio_for_host_B.wav" },
    "3": { "reference_audio": "path/or/url_to_user_audio_for_host_A.wav" }
  }
}

请根据生成的SRT中Host A和Host B对应的台词调整片段编号。

3. 渲染音频:

bash

undefined

If using user's reference_audio, use the noiz backend

bash path/to/tts.sh render --srt podcast.srt --voice-map voice_map.json --backend noiz -o podcast_output.mp3

undefined

bash path/to/tts.sh render --srt podcast.srt --voice-map voice_map.json --backend noiz -o podcast_output.mp3

undefined

Step 5: Present the Result

步骤5：展示结果

Let the user know that the podcast has been successfully generated. You MUST provide both files back to the user:

Show or link to the
```
podcast_script.md
```
file so they can read the script.
Provide the path to the
```
podcast_output.mp3
```
file so they can listen to the audio.
Briefly summarize the headlines that were included in the podcast.

告知用户播客已成功生成。你必须向用户提供以下两个文件：

展示
```
podcast_script.md
```
文件或提供其链接，方便用户查看脚本。
提供
```
podcast_output.mp3
```
文件的路径，方便用户收听音频。
简要总结播客中包含的新闻头条。