Public Feed Discovery & Latest Items Agent

Find public RSS, Atom, and JSON Feed endpoints for a site or direct feed URL, then return a bounded latest-items brief with source feed URLs, publication dates, freshness metrics, focus-term matches, evidence URLs, confidence, and diagnostics.

This Actor is for content operations, SEO, market monitoring, PR, newsletter research, and AI ingestion planning when you need to know whether a public site has usable feed data before building a larger crawl or article pipeline.

What It Returns

• Sanitized input URL and site origin
• Same-site feed URLs found from public HTML or common feed paths
• Candidate feed evidence showing direct input, HTML discovery, or common-path probing plus readable/fetch/parse status
• RSS, Atom, or JSON Feed summaries
• Latest feed items with title, sanitized item URL, published date, author, summary snippet, source feed URL, and matched focus terms
• Freshness summary with date coverage and recent item counts
• Feed discovery hash
• Evidence URLs
• Confidence, completeness, missing fields, diagnostics, and readable errors

Pricing Design

The intended pay-per-event setup is:

• apify-actor-start: a tiny run-start fee
• useful-feed-discovery-result: charged only for useful public feed discovery records with at least one readable feed and one item
• no apify-default-dataset-item

If focusTerms are supplied, the useful event is charged only when at least one feed item matches a focus term. Invalid inputs, duplicates, private-network URLs, empty feeds, failed resources, and focus-term misses should not charge the useful event.

Good Fits

• Discovering whether a public website exposes RSS, Atom, or JSON Feed data
• Building a source list for content monitoring
• Finding recent public posts around a focus topic
• Preparing feed-backed inputs for AI/RAG workflows
• Auditing feed freshness and date coverage

Boundaries

This Actor does not crawl full articles, log in, use cookies, render JavaScript, scrape search engines, scrape social platforms, bypass paywalls, or read private feeds. It fetches only public same-site HTML and feed resources. Credentials, query parameters, fragments, path params, private-network addresses, localhost, .local, sensitive reset/invite paths, and token-like paths are rejected, skipped, or safely redacted. Item URLs are sanitized and are not fetched.