Extractor — an Ounie app for AI agents

Any document in, clean text out. Point Extractor at a URL or hand it a file — PDF, Word, PowerPoint, Excel, HTML, Markdown, CSV, or an image — and get back clean markdown, plain text, or JSON in the exact shape you asked for.

One call, one document: 2 credits. Long PDFs bill 2 credits per 50-page block, counted on pages that actually carried text
Strips the packaging: navigation, cookie banners, newsletter modals, footers and legal boilerplate come off; headings, lists, tables and code stay
format: "json" fills only fields the document actually states — anything unstated comes back null and is listed in missing_fields, never invented
Credits are reserved before the work and settled on the document's real size; a source with no extractable text refunds in full and is never charged
Built for AI agents day one: xtr_live_ REST keys, an MCP server (works with the Ounie AI Team via ?api_key=), keyless x402 pay-per-call in USDC on Base, and the fleet master key
Audio and video are out of scope by design — those are billed per minute at transcribe.ounie.com

For agents

Works with Claude, ChatGPT, and any MCP client. One server, every app: paste the gateway config and the agent can discover and run Extractor — and the rest of the store — through ounie_list_apps → ounie_describe_app → ounie_invoke_app.

{
  "mcpServers": {
    "ounie": {
      "url": "https://mcp.ounie.com"
    }
  }
}

Scoped endpoint — just this app. Point a client here when the agent should only ever see Extractor's tools:

{
  "mcpServers": {
    "extract": {
      "url": "https://extract.ounie.com/api/mcp"
    }
  }
}

To spend your Ounie credits on the scoped endpoint, append your app key to the URL as ?api_key=… — the header-less form clients like the Ounie AI Team need.

For agents

Credits used