Parse RSS, Atom, and JSON feeds and discover feeds behind page URLs.
npx skills add https://github.com/Chia1104/agent-air --skill rss-feeds-chia1104
Please help me install this Agent Skill. Skill: rss-feeds Source: https://github.com/Chia1104/agent-air/tree/main/skills/hermes/research/rss-feeds Command: npx skills add https://github.com/Chia1104/agent-air --skill rss-feeds-chia1104
<link rel="alternate"> tags or common paths like /feed and /rss.xml. - Filtering & Automation: Bound results with --limit or --since for incremental cron-based digests, and emit machine-readable JSON with --json. - Use Case: Ask for the latest releases of a GitHub repository; the Skill reads https://github.com/OWNER/REPO/releases.atom and returns a dated list of release entries with links you can cite in a report. ## Quick Start Ask the agent to read the latest 10 entries from a blog or site URL, for example: "Show me what's new on simonwillison.net using the rss-feeds skill."High-intent search queries and answers about installing and using this skill.
Run `python3 scripts/feed.py read <feed-url> --limit 10` to print a date-sorted list of entries with title, link, date, author, and summary. Add `--json` for machine-readable output or `--since YYYY-MM-DD` to keep only newer entries.
Use the discover command: `python3 scripts/feed.py discover https://example.com/`. It scans the page's `<link rel="alternate">` tags first, then falls back to common paths like /feed, /rss.xml, and /atom.xml. The read command also discovers automatically when given a page URL.
No. The script uses only the Python 3.10+ standard library (urllib, xml.etree, json), so it runs in any environment without pip installs. It needs network access to the feed host.
No, this Skill returns only feed metadata: title, link, published date, author, and a summary capped at 2000 characters. Feed summaries are often truncated, so pass an entry's link to a web extraction tool for the full article body.
A 403 usually means the feed is behind Cloudflare or blocks non-browser clients. A 'no feed found' result means the page has no link tags and none of the common feed paths exist; check the site footer or /sitemap.xml before concluding no feed exists.
Persist the last-seen published date from each run and pass it as `--since` on the next run, so only new entries are returned. Feeds are cheaper and more stable than scraping HTML front pages on every scheduled run.
taracodlabsMonitor RSS and Atom feeds for new posts using Python feedparser.
loteironParse RSS, Atom, and JSON feeds and discover feeds behind page URLs.
Monjyu1101Track RSS/Atom feeds and manage read/unread articles via blogwatcher-cli.
rbudiharsoMonitor blogs and RSS/Atom feeds using the blogwatcher-cli command-line tool.