RSS & Atom Feed Monitor: Only New Items

RSS & Atom Feed Monitor: Only New Items, Keyword Filters, Excel

Use this to watch many RSS, Atom or JSON feeds and get only the items that are new since your last run. Paste feed URLs, or just website and blog URLs: the feed is found automatically. Each row has the feed, title, link, publish date and age, author, plain-text summary, categories, podcast/media enclosure and the keywords it matched. Filter by keywords, excluded words and maximum age. $0.0005 per item.

• 🔁 Only new items: the Actor remembers what it already returned (per state key), so a scheduled run returns and charges only fresh items
• 🔎 Feed autodiscovery: give https://blog.example.com and it finds /feed, /rss.xml or the <link rel="alternate"> feed
• 🧰 RSS 2.0, Atom and JSON Feed, up to 500 feeds per run, fetched in parallel
• 🏷️ Keyword include/exclude across title, summary and categories, plus maximum age in hours
• 🩺 FEED_STATUS record: which feeds worked, which failed and why (failed feeds are free)
• 💸 $0.0005 per item; runs with nothing new are free

Who it's for

• Marketing and PR teams: competitor blogs, press rooms and industry news in one daily sheet
• Sales and RevOps: trigger events (funding, launches, hiring posts) from target-company blogs
• Analysts, compliance and policy teams: regulator, agency and court RSS feeds filtered by keyword
• Newsletter writers and content curators: a clean daily digest with summaries and links
• AI agents and automations: a reliable "what's new" step before summarising or posting

How to use it

1. Click Try for free.
2. Paste feeds or websites (one per line).
3. Optional: keywords, exclude keywords, maximum age (hours), a state key for this watch list.
4. Click Start, then schedule it (hourly or daily). Download RSS_XLSX or RSS_CSV from Storage → Key-value store, or connect the dataset to Slack, email or Sheets.

Input example

{
  "feeds": ["https://blog.apify.com", "https://github.blog", "https://www.theverge.com"],
  "keywords": ["AI", "agent", "scraping"],
  "maxAgeHours": 48,
  "onlyNew": true,
  "stateKey": "tech-watch"
}

Task examples

• Competitor blog watch: {"feeds": ["competitor1.com/blog", "competitor2.com/news"], "stateKey": "competitors"} daily
• Regulator alerts with keywords: {"feeds": ["https://www.federalregister.gov/api/v1/documents.rss?conditions%5Bagencies%5D%5B%5D=food-and-drug-administration"], "keywords": ["device", "recall"]}
• Hacker News stories over 100 points about AI: {"feeds": ["https://hnrss.org/newest?points=100"], "keywords": ["AI", "LLM"], "maxAgeHours": 24}
• Podcast episode tracker: {"feeds": ["https://feeds.example.com/show.xml"], "onlyNew": true} and read enclosureUrl
• One-off digest of the last week, no memory: {"feeds": [...], "maxAgeHours": 168, "onlyNew": false}

Output example

{
  "feedTitle": "The GitHub Blog",
  "feedUrl": "https://github.blog/feed/",
  "siteUrl": "https://github.blog/",
  "title": "Beyond code generation: A developer’s guide to agentic engineering",
  "link": "https://github.blog/ai-and-ml/github-copilot/beyond-code-generation-a-developers-guide-to-agentic-engineering/",
  "published": "2026-11-06T22:15:59+00:00",
  "ageHours": 65.2,
  "author": "Nick Liffen",
  "summary": "Writing code got faster. The work around it didn't. See how you can plan, write code, review changes, and investigate production issues with GitHub Copilot, and what work still needs your judgment. The post Beyond code generation: A developer’s guide to agentic engineering appeared first on The GitHub Blog .",
  "categories": [
    "AI & ML",
    "GitHub Copilot",
    "GitHub Universe"
  ],
  "enclosureUrl": null,
  "matchedKeywords": [],
  "itemKey": "392d1ff82a7bb255fe2e",
  "agentSummary": "The GitHub Blog: \"Beyond code generation: A developer’s guide to agentic engineering\" 2026-11-06 (65.2h ago) — Writing code got faster. The work around it didn't. See how you can plan, write code, review changes, and investigate production issues with GitHub Copilot, an… https://github.blog/ai-and-ml/github-copilot/beyond-code-generation-a-developers-guide-to-agentic-engineering/"
}

The item above is real output from a test run on 2026-11-09.

For AI agents

Use this Actor to answer "what's new on these sites or feeds since last time, about topic X?". Minimal input: {"feeds": ["https://github.blog"], "maxAgeHours": 24}. Read agentSummary per item (feed, title, date and age, short summary, matched keywords, link) or the SUMMARY takeaway (count per feed and the newest headline). Check FEED_STATUS for feeds that failed.

• MCP: add this Actor to Claude, Cursor, Copilot or any MCP client through the Apify MCP server: https://mcp.apify.com/?actors=pandadad0214/rss-feed-monitor, then ask in plain English.
• API: POST https://api.apify.com/v2/acts/pandadad0214~rss-feed-monitor/run-sync-get-dataset-items?token=YOUR_TOKEN with the input JSON.
• Keep a separate stateKey per watch list so "only new" memory doesn't mix between agents or clients.

Pricing

$0.0005 per feed item returned (pay per event) plus Apify's tiny Actor-start fee of $0.00005. 1,000 items cost $0.50; a daily watch of 20 feeds returning ~60 new items costs about $0.03 a day. Runs with nothing new cost nothing, and failed or blocked feeds are never charged. Set Maximum results or Apify's maximum charge per run to cap any run.

Apify paid-plan discounts: Bronze members pay 10% less, Silver 20% less and Gold 30% less per item; the discount applies automatically. Apify's free plan includes monthly credit to try it.

FAQ

How does "only new" work? Each returned item's ID (GUID, link or title+date) is stored in a named key-value store in your own Apify account (pf-rss-feed-monitor-state), under your state key. Next run, those items are skipped and not charged. Use a new state key, or set onlyNew to false, to start fresh.

What if a site has no feed? The run lists it in FEED_STATUS as "no RSS/Atom feed found" at no charge. Use the Website Change Monitor for pages without feeds.

Does it read full articles? No. It returns what the feed publishes (title, summary or content, link). That keeps it fast, cheap and polite to publishers.

Is it safe to point at any URL? Only public http(s) addresses are fetched. Private, internal and cloud-metadata addresses are blocked, including after redirects.

Part of Panda Federation

This Actor is part of the Panda Federation catalog of public-data tools. See the rest of the work on the official site.

More Panda Federation tools

• Website Change Monitor: watch pages that have no feed
• Federal Register Monitor: new US federal rules and comment deadlines
• Podcast Leads Finder: find podcasts and hosts by topic
• Broken Link Checker: find dead links on a site
• Full catalog and MCP server: Panda Federation Tools on GitHub

Built by Panda Federation. Questions or a custom version? Open an issue on the Actor's Issues tab.

Changelog

• 2026-11-10: Store copy polish (SEO title and description, working example input). README now links the official Panda Federation site with listing campaign tags. Pricing unchanged.
• 2026-11-09: First release (0.1.0): RSS 2.0, Atom and JSON Feed parsing, website feed autodiscovery, only-new memory per state key, keyword include/exclude, maximum age, per-feed item cap, FEED_STATUS health record, SSRF-safe fetching, Excel and CSV (RSS_XLSX / RSS_CSV), agentSummary and SUMMARY takeaway, pay-per-event pricing with Bronze/Silver/Gold discounts.