Article Content Extractor — Clean Text for RAG & LLM

What you get: Clean article body, byline, publish date, excerpt, and optional hero image for RAG/LLM and content QA. Export Markdown/CSV. No login.

Batch audit (opt-in): Set generateReport: true to add one article-content-audit-report row that summarizes the batch (article count, full vs partial, word totals, confidence, per-URL warnings). It charges $2.50 once per run when at least one article is usable; leave it off for the first ≈$0.008 proof.

Sample output (2 real rows from recent default-input runs, 2026-10-13):

title	siteName	publishedDate	wordCount
New touch-up tools in Google Photos’ image editor let you make quick,…	Google	2026-04-20	123
Web scraping - Wikipedia	Wikimedia Foundation, Inc.	2005-09-17	4401

Use it for:

• RAG ingestion: turn article URLs into clean Markdown with title, byline and date.
• Content QA: check that published posts have a title, description and publish date.
• Research digests: collect clean text from news and blog posts for summarising.

CTA: Open Input (link below) → leave the prefilled Wikipedia URL → Start. Leave report/export OFF. First paid proof ≈ $0.008. From $8/1k useful rows. Open Input (leave defaults → Start)

Paid first run ≈ USD 0.008 total (1 article URL): useful row USD 0.008. Leave includeImages, generateReport, and emitExport OFF for first proof. Open Console Input (deep link above) and run this JSON.

{
  "urls": [
    "https://en.wikipedia.org/wiki/Web_scraping"
  ],
  "outputFormat": "markdown",
  "includeImages": false,
  "generateReport": false,
  "emitExport": false,
  "delivery": "dataset",
  "dryRun": false
}

Paste 1 article URL and Start. Charged $0.008 per useful row only.

Schedule example (recurring runs)

Create an Apify Schedule that starts this Actor on your cadence (daily/weekly). Keep the same URL list and leave the expensive flags off until the first few runs look right. When you want a per-run summary for the batch, set generateReport: true (still opt-in; default off):

{
  "urls": [
    "https://en.wikipedia.org/wiki/Web_scraping",
    "https://en.wikipedia.org/wiki/Web_crawler"
  ],
  "generateReport": true,
  "emitExport": false,
  "delivery": "dataset",
  "dryRun": false
}

Cost: the article rows as usual ($0.008 each) plus one article-content-audit-report row ($2.50 + $0.008 row) per run, only when at least one article is usable. On a schedule this is charged on every run, so use it for batches where a summary is worth it, and set Max cost per run to your budget. Both generateReport and emitExport stay OFF by default.

High-value example (batch export or Site QA handoff)

For a buyer-facing deliverable on top of the cheap article rows, pick one:

1. In-actor export handoff — set emitExport: true for one article-batch-export row ($5.00 + $0.008 row) when at least one article is usable:

{
  "urls": [
    "https://en.wikipedia.org/wiki/Web_scraping",
    "https://en.wikipedia.org/wiki/Web_crawler"
  ],
  "generateReport": false,
  "emitExport": true,
  "delivery": "dataset",
  "dryRun": false
}

2. Linked Site QA / RAG report — keep this Actor as the low-cost extraction step, then send the same URLs to Site QA Content Report Scraper, Site QA Indexability AI Crawler Report Scraper, or Website RAG Readiness Audit. Keep extraction and the report as separate runs so the first cost and the high-value deliverable stay clear.

Store Quickstart

Run the low-cost extraction with a single authorized URL; leave images and paid reports OFF (~$0.008):

{
  "urls": ["https://en.wikipedia.org/wiki/Web_scraping"],
  "includeImages": false,
  "generateReport": false,
  "emitExport": false,
  "delivery": "dataset",
  "dryRun": false
}

Use with Apify MCP

Exact actor: taroyamada/article-content-extractor

Sample input JSON:

{
  "urls": [
    "https://en.wikipedia.org/wiki/Web_scraping"
  ],
  "outputFormat": "markdown",
  "includeImages": false,
  "generateReport": false,
  "emitExport": false,
  "delivery": "dataset",
  "dryRun": false
}

Run the next report

Turn a small extraction into a decision-ready Site QA workflow:

• Run a content QA report for source-linked title, metadata, thin-content, and page-quality issues.

• Run an indexability and AI crawler report for robots, sitemap, canonical, noindex, schema, and crawler-access findings.

• Run a website RAG readiness report for retrieval structure, coverage, and source-linked readiness findings.

This follow-on actor uses user-supplied public URLs only. It does not provide ranking guarantees, legal advice, or automated messaging workflows.

Content teams and QA teams provide public article URLs they own or are allowed to audit. Use the extracted URLs as the source step for a Site QA report or export, then review source-linked issues rather than raw HTML alone.

The report actors below produce the buyer-facing deliverable; this actor remains the bounded extraction step.

High-value Site QA/RAG handoff

See High-value example above for the emitExport JSON and linked report Actors. Use this actor as the low-cost source step, then send the same URLs to one linked Site QA or RAG report actor. Keep extraction and the buyer-facing report as separate runs so the first cost and the high-value deliverable are clear.

Input examples

Article URLs (images ON — opt-in)

{
  "urls": ["https://en.wikipedia.org/wiki/Web_scraping"],
  "includeImages": true,
  "delivery": "dataset",
  "dryRun": false
}

Press pages

{
  "urls": ["https://example.com/press/release"],
  "includeImages": false,
  "delivery": "dataset",
  "dryRun": false
}

Value path output preview

The report/export output is the buyer-facing result. The extraction row below is only the source step:

{
  "actorName": "site-qa-content-report-scraper",
  "rowType": "site_qa_report",
  "billingEventName": "site-qa-report-generated",
  "sourceUrl": "https://example.com/news/example",
  "issueCount": 2,
  "reportStatus": "action_needed"
}

Article value events

Use generateReport=true to emit one article-content-audit-report row at $2.50, or emitExport=true to emit one article-batch-export row at $5.00, when at least one article is usable. These are the buyer-facing batch deliverables; raw article rows remain source context.

Output fields

• rowType
• url
• headline
• author
• publishedAt
• articleText
• excerpt
• heroImage
• sourceUrl

Rows also include source URLs, fetch timestamps, warnings when a source is partial, and stable IDs when the workflow supports recurring change detection.

Pricing and no-change runs

$0.00005 actor start, $0.008 per useful article row, $2.50 per article content audit report, and $5.00 per batch export. Failed/no-content rows stay out of the default dataset.

The default dataset is the canonical billable surface. Dry runs, validation-only runs, error-only runs, and no-usable-row recurring polls do not write payable rows. A webhook is sent only after dataset billing succeeds and is best effort; a failed webhook is not retried and never creates a second charge.

Request limits are bounded at 300 URLs, 3 redirects, and 5 MiB per HTML response by default. maxRedirects and maxResponseBytes can be lowered for stricter runs; the runtime never permits more than 10 redirects or 10 MiB.

Compliance guardrails

• Fetch public article pages supplied by the user.

• Do not imply content ownership transfer or publisher endorsement.

• Use output for research, QA, and internal workflows.

• Do not use provider emblems or wording that implies approval by an upstream data provider.

• Only fetch public HTTP(S) URLs that you own or are authorized to audit. Loopback, private, link-local, metadata, reserved, and DNS-resolved private targets are rejected, as are non-HTML responses, excessive redirects, and oversized responses.

See Also

The three report paths in Run the next report are the complete Site QA/RAG handoff for this actor. Use one of those paths instead of adding more raw extraction volume.

💾 Save it for later: click the bookmark icon at the top of the Apify Store page if you'd like to come back to it. Bookmarks help other engineers find this actor via Apify's discovery surfaces.

Feedback

Use the Apify Store rating and issue controls on the actor page for feedback. The three report paths above remain the only cross-actor workflow links.

Related Actors

• taroyamada/website-content-extractor
• taroyamada/rss-feed-aggregator
• taroyamada/google-news-scraper
• taroyamada/broken-link-checker
• taroyamada/meta-tag-analyzer

Works well with

• Apple Podcasts Scraper — search shows, look up IDs, pull recent episodes and listener reviews
• Trade Show Exhibitor Intelligence — turn exhibitor pages into booth and lead lists
• Apple Podcasts Chart Tracker — track top-chart ranks by country/genre with movement reports