This skill enables an agent to discover, fetch, and read content from online RSS and OPML feeds. It is designed to be operated via command-line execution.
The execution environment must support:
- Python 3.x.
- The ability to install Python packages from a
requirements.txtfile. - Network access to fetch remote content.
The required Python packages are listed in scripts/requirements.txt.
Before first use, a configuration file must be set up to specify the source of the RSS feeds.
- File to Edit:
assets/config.json - Action: Provide a valid URL for an OPML file in the
opml_urlfield.
Example config.json:
{
"opml_url": "https://your-opml-provider.com/feeds.opml"
}Note: The assets directory is used for local configuration and caching. It should be excluded from version control.
The skill's functionality is exposed through the scripts/rss_fetcher.py script.
Action: Discovers all available RSS feeds from the configured OPML URL. Syntax:
python scripts/rss_fetcher.py --opml-url <your-opml-url>
Action: Fetches the list of recent articles from a specified RSS feed. Syntax:
python scripts/rss_fetcher.py --feed-url "<feed_xml_url>" --name "<feed_name>"
<feed_xml_url>: The direct URL to the feed's XML file.<feed_name>: A unique name for the feed, used for caching.
Action: Fetches the full content of a specific article. Syntax:
python scripts/rss_fetcher.py --content-url "<article_url>" --title "<article_title>" --author "<article_author>" --date "<publication_date>"
<article_url>: The URL of the article to read.- Other parameters are used for creating a unique cache entry.