RSS & Atom Feed to JSON

Turn any RSS, Atom or RDF feed into clean, normalized JSON: one dataset item per article with ISO 8601 dates, plain-text and HTML content, parsed author names, categories and podcast enclosures — the same field names no matter which feed format the source uses. Multiple feeds per run; broken feeds are reported as failed items instead of crashing the batch. No credentials needed.

What it does

Give it a list of feed URLs. For each article you get one dataset item with:

• title, url, id — GUID (RSS) / id (Atom), falling back to the article URL
• published, updated — normalized to UTC ISO 8601 (2026-08-17T09:30:00Z), any input format
• author — parsed to a human name (RSS email (Name) convention handled)
• summaryText, contentText — HTML stripped to clean plain text (scripts/styles removed)
• contentHtml — the original (sanitized) HTML content, when you want the markup
• categories — tag/category terms as a string array
• enclosures — podcast audio / media attachments as {url, type, lengthBytes}
• feedUrl, feedTitle, feedType (rss20/atom10/…), language — the source feed

Feeds that fail to download or parse produce a single item with status: "failed" and a clear error message, so batch jobs always finish.

Input

Field	Type	Default	Description
feedUrls	array	—	Feed URLs (redirects followed)
maxItemsPerFeed	integer	0 (all)	Keep only the first N articles per feed
includeContentHtml	boolean	true	Include contentHtml alongside contentText
timeoutSecs	integer	30	Download timeout per feed
maxFileSizeMb	integer	10	Larger feeds are skipped with a failed item

Example output (abridged)

{
    "feedUrl": "https://fruit.example.com/feed.xml",
    "feedTitle": "Fruit Weekly",
    "feedType": "rss20",
    "status": "success",
    "id": "fruit-weekly-001",
    "url": "https://fruit.example.com/posts/apples",
    "title": "Apple shipments & prices rise",
    "author": "Ann Orchard",
    "published": "2026-08-10T08:30:00Z",
    "summaryText": "Apple volumes grew 12% this week.",
    "contentText": "Apple report\nApple volumes grew 12% this week.\nFuji up\nGala flat",
    "contentHtml": "<h2>Apple report</h2><p>Apple volumes grew <b>12%</b> this week.</p>",
    "categories": ["apples", "markets"],
    "enclosures": [],
    "language": "en-us"
}

Typical uses

• News monitoring and research pipelines (feeds → LLM/RAG-ready text)
• Aggregating many blogs/podcasts into one normalized dataset
• Feeding no-code tools (Zapier/Make via Apify integrations) with clean article JSON

Limitations

• Content comes from the feed itself — articles are not fetched from their web pages (feeds that only carry teasers yield teaser-length contentText)
• Password-protected or non-public feeds are not supported

Development (local)

cd actors/rss-atom-to-json
uv venv --python 3.13 .venv && uv pip install -p .venv/bin/python -r requirements.txt

.venv/bin/python tests/run_local_test.py           # end-to-end test (apify run equivalent), exit 0 = ALL PASS
../../node_modules/.bin/apify run                  # real apify CLI local run (input: storage/key_value_stores/default/INPUT.json)

Test fixtures are tests/fixtures/sample_rss.xml / sample_atom.xml / not_a_feed.html; expected results live in tests/expected_output.json. Publishing → ../../docs/publishing.md.