Pricing

from $0.70 / 1,000 fetched pages

Try for free

Go to Apify Store

Agent Fetch: URL to Markdown for LLMs and RAG

Try for free

Returns the main content of any web page URL as clean Markdown for LLMs, RAG and AI agents, with title, canonical URL, language, word count and links. Up to 25 URLs per run, automatic JavaScript rendering, optional JSON Schema field extraction with no LLM. Failed pages are free.

Pricing

from $0.70 / 1,000 fetched pages

Rating

0.0

(0)

Developer

Omar Nagy

Actor stats

Bookmarked

Total users

Monthly active users

a month ago

Last modified

Categories

Agents

Developer tools

Security

Requested URL

url

Optional

The URL exactly as you passed it in, so results can be joined back to the input list.

Type:string

Example:

https://www.propublica.org/article/3m-forever-chemicals-pfas-pfos-inside-story

Final URL

finalUrl

Optional

The URL actually fetched after redirects.

Type:string | null

Example:

https://www.propublica.org/article/3m-forever-chemicals-pfas-pfos-inside-story

Status

status

Optional

Outcome of this page. ok means markdown was produced. Every other value is a labelled failure, so a failed page never silently looks empty. needs_render means the plain fetch got an empty JavaScript app shell and autoRender was off.

Type:string

Example:

HTTP status

httpStatus

Optional

HTTP response code of the final request. 0 when no response was received at all.

Type:number | null

Example:

200

Method

method

Optional

How the page was retrieved: http (plain fetch) or browser (headless Chromium, set by render or autoRender).

Type:string | null

Example:

http

Fetched at

fetchedAt

Optional

ISO 8601 timestamp of when the page was retrieved.

Type:string | null

Example:

2026-09-04T22:50:15.537Z

Title

title

Optional

The page's title.

Type:string | null

Example:

Toxic Gaslighting: How 3M Executives Convinced a Scientist the Forever Chemicals She Found in Human Blood Were Safe

Canonical URL

canonicalUrl

Optional

The canonical URL the page declares for itself, useful for de-duplicating syndicated content.

Type:string | null

Example:

https://www.propublica.org/article/3m-forever-chemicals-pfas-pfos-inside-story

Language

language

Optional

Language tag the page declares.

Type:string | null

Example:

en-us

Markdown

markdown

Optional

The main content of the page as clean Markdown, with navigation, ads and boilerplate removed. This is the field to feed an LLM.

Type:string | null

Example:

Kris Hansen had worked as a chemist at the 3M Corporation ...

Words

wordCount

Optional

Word count of the extracted Markdown.

Type:number | null

Example:

Characters

charCount

Optional

Character count of the extracted Markdown, after any cut by maxMarkdownChars. Never more than maxMarkdownChars.

Type:number | null

Example:

397

Links

linkCount

Optional

How many links the main content contains.

Type:number | null

Example:

Truncated

truncated

Optional

Whether the Markdown was cut short by your maxMarkdownChars limit. A cut Markdown ends with [truncated by maxMarkdownChars].

Type:boolean | null

Example:

true

Duration

durationMs

Optional

Milliseconds from starting this URL to its record, including any wait for the converter.

Type:number | null

Example:

638

Reason

reason

Optional

Present on failed pages. Plain-language explanation of why the page did not convert, for example blocked by robots.txt.

Type:string | null

Example:

Disallowed by robots.txt

Links

links

Optional

Present when includeLinks is on: every link inside the extracted content, as { url, text, external }. Absolute URLs, capped at 500 per page.

Type:array | null

Example:

[object Object]

Article metadata

article

Optional

Present when includeArticleJson is on: excerpt, byline, site name, published and modified dates, description, lead image and JSON-LD blocks. Only keys with a value are included.

Type:object | null

Extracted fields

extractedFieldCount

Optional

How many of your schema fields were found on the page. Present only when a schema was supplied.

Type:number | null

Example:

Extracted values

extracted

Optional

The values for the fields your schema asked for, keyed by field name. Deterministic parsing, no LLM and no API key.

Type:object | null

Extraction sources

extractedSources

Optional

Where each extracted value came from, so a value can be audited. json-ld, meta, page:pairs and text:label are values the page itself states; page:title, page:canonical, page:lang, page:wordCount, html:time and text:pattern are fallbacks and are never charged.

Type:object | null

Rendered automatically

autoRendered

Optional

Present and true when the plain fetch returned an empty JavaScript app shell and the page was loaded in the headless browser instead (autoRender). Such a page is charged as a render.

Type:boolean | null

Website & Page to Markdown for RAG and LLMs

humble-echidna/page-to-markdown

Turn web pages or whole sites into clean Markdown for RAG and LLMs: main content only, with title, links and word count. Whole-site mode reads the sitemap first, can return only changed pages, and splits pages into heading-based chunks. Respects robots.txt and AI opt-outs. No browser.

Michael Costa

Website to Markdown Converter & Crawler for LLMs and RAG

s3nafps/ai-ready-webpage-extractor

Website to Markdown Crawler is an Apify Actor that turns web pages, docs sites and blogs into clean Markdown and RAG chunks for LLMs. $3 per 1,000 pages.

mohamed senator

Website to Markdown for LLM & RAG — Crawl4AI URL to Clean

bikram07/web-to-markdown-crawl4ai

Convert any URL, sitemap, or whole website into clean, LLM-ready Markdown for RAG, vector databases, and AI agents. Hosted Crawl4AI in a real Chromium browser — renders JavaScript and SPAs, strips boilerplate, and exports JSON/CSV. Callable over MCP from Claude and Cursor.

Bikram

Site to Agent Feed (URL to RAG-ready Markdown)

constant_quadruped/site-to-agent-feed

Turn any URL into clean, RAG-ready Markdown + structured JSON for LLMs and AI agents. Self-healing main-content extraction (survives redesigns), headings/links/tables, optional change-detection. No paid APIs.

URL to Clean Markdown — for AI Agents

great_saint/url-to-markdown

Convert any web page into clean, LLM-ready Markdown. Charged only on successful conversion. MCP & x402 ready.

Öge

Web Fetch — Any URL to Markdown or Text for AI Agents

actor-forge/web-fetch

Fetch any URL as clean Markdown and plain text for AI agents, MCP and RAG: web pages, PDF, Word, Excel, PowerPoint, EPUB, CSV and JSON. One url in, one result with title and token estimate out. Static HTML, no JavaScript rendering. $1 per 1,000 pages.

ActorForge

Website to Markdown for LLMs & RAG — Content Extractor

fetchbase/website-to-markdown

Turn any URL or whole site into clean, LLM-ready Markdown, text, or JSON. Strips nav/ads/boilerplate; keeps headings, links, tables, code. Sitemap-aware concurrent crawl, URL filters, robots.txt respected, rich metadata. Charged only per page extracted — no startup fee.

Fetchbase

Website Content Extractor — Pages to Markdown for LLM/RAG

taroyamada/website-content-extractor

Turn docs, help centers and pricing pages into clean Markdown for LLM and RAG: one row per URL with main text, title, word count and a content-quality score. No API key or login.

naoki anzai

Web Page to Markdown Converter — Clean LLM Content

maged120/url-to-markdown-pro

Convert web pages to clean Markdown for LLMs, or extract just the article text (reader view) as plain text, HTML or Markdown with title, headings and word count. Batch URLs, no setup.

Maged

URL to Clean Markdown for LLMs & RAG

eltociear/url-to-markdown-for-llms

Convert a webpage to Markdown and extract the main content from any URL. Turns HTML into clean Markdown for LLM context and RAG ingestion: strips nav, ads and boilerplate, keeps headings, lists and tables, returns title, byline, date and word count. No headless browser. Pay per page extracted.