Ai Tech Rss Fetch

What it does

This skill, Ai Tech Rss Fetch, allows an AI agent to subscribe to RSS/Atom feeds from URLs or OPML files and persistently store their metadata in a SQLite database. It focuses on reliably fetching and deduplicating feed entries without retrieving the full article content or generating summaries. The primary function is to maintain a stable record of feed information for downstream processing.

When to use it

• You need to track updates from multiple tech news sources.
• You require a reliable, persistent store of RSS feed metadata.
• You want to avoid fetching full article content and generating summaries as part of the initial data collection process.
• You're building an AI agent that requires access to recent tech news headlines and basic information without needing the complete articles.
• You need a system for deduplicating RSS feed entries across multiple runs.

Key capabilities

• Subscribes to RSS/Atom feeds from URLs or OPML files.
• Persists feed metadata (feed URL, title, site URL, etc.) and entry metadata (title, author, published date, summary hash, etc.) to a SQLite database.
• Deduplicates entries using layered identity keys and content fingerprints.
• Supports incremental synchronization of feeds.
• Provides configurable parameters for database path, feed URLs, maximum number of feeds/items per run, and more.

Example prompts

• "Subscribe to the RSS feed at https://example.com/feed.xml."
• "Import the tech blogs from this OPML file: assets/hn-popular-blogs-2025.opml."
• "Run a sync of all subscribed feeds, limiting it to 20 feeds with 100 items per feed."

Tips & gotchas

• Database Path: In multi-agent environments, ensure you explicitly set the AI_RSS_DB_PATH environment variable to an absolute path before running any commands.
• No Summaries: This skill does not fetch full article bodies or generate summaries. It only stores metadata.
• Dependency Installation: Requires feedparser Python package: python3 -m pip install feedparser.