MH MediaHarvester Web Context API
One URL → web data, media, brand and image workflows

Harvest the web.Keep the useful parts.

Paste a public URL or domain and run a real MediaHarvester request. Scrape clean web context, extract media, build a live brand profile or capture a screenshot from the same workspace.

Real requests use the same authenticated API endpoints documented for developers.
Try it View API
Ready for a live request.
GET /v1/web/scrape/markdown ready
Web Data · Scrape example.com
↗ Run a live request

The result panel will show data returned by the selected MediaHarvester endpoint — not a scripted demo.

25documented data tools
5SDK languages in this repo
88conversion workflows
REST + CLI + MCPdeveloper interfaces shipped

Live workspace

Build an extraction request

Public targets · no extension
Request builder Simple mode URL + media type + run. Open Advanced when you need crawl, Drive, dedupe or size rules.

Scope

Target URLs or folder links

Source

Where should we collect from?

Mode

What should we collect?

Output options

Duplicate handling

Crawl strategy

Pagination, infinite and branching behavior

Execution

Preview and export

POST /api/generate-download
Source Site
Selected mode Images
Duplicates Allow repeats
Pagination Follow until last page
Infinite Load next 5 levels
Branch depth Do not branch
Behavior

Images are downloaded directly. Videos are exported as discovered source links. Files are downloaded as filtered assets. Google Drive mode reads public folder inventory directly instead of scraping rendered Drive page visuals.

Source behavior

Site mode explores accessible pages and discovered media. Google Drive mode reads the contents of public folders and uses public download links.

Branching model

In site mode, internal links can be followed up to 3 levels. In Google Drive mode, a dedicated subfolder depth control recursively expands public folders up to 3 levels.

API example
{
  "source_mode": "google_drive",
  "mode": "files",
  "dedupe_media": true,
  "file_types": ["pdf", "docx", "css"],
  "file_size_filter": "greater_than",
  "file_size_kb": 500,
  "drive_subfolder_depth": 1,
  "pagination_mode": "to_end",
  "infinite_scroll_mode": "next_5",
  "link_branch_depth": 1,
  "urls": ["https://drive.google.com/drive/folders/your-public-folder-id"]
}
Sources Sites + public Google Drive folders
Discovery Classic crawl + direct Drive inventory
Output ZIP, CSV, TXT, manifest