Demo: PDF → agent-readable text

Flow

INPUT → discovery → price → 402 (if unpaid) → settle/prepaid → OUTPUT.

{"file_base64":"<pdf-base64>","options":{"includeTables":true,"maxPages":5}}

{
  "pages": [
    {
      "page": 1,
      "text": "Example PDF\nSection 1\nHello from page one."
    }
  ],
  "tables": [
    {
      "page": 1,
      "rows": [
        [
          "Item",
          "Qty"
        ],
        [
          "Widgets",
          3
        ]
      ]
    }
  ],
  "meta": {
    "pageCount": 1,
    "bytesDecoded": 12345
  }
}

Latency

Depends on PDF size and page count (decoded ≤10 MiB). End-to-end includes network + billing.

Try it

curl -sS https://api.callmeter.dev/v1/agentutils/pdf/extract \
  -H "X-Api-Key: $CALLMETER_API_KEY" \
  -H "Content-Type: application/json" \
  -H "X-BizKernel-Acquisition: qa_agent_pdf_demo_01" \
  -d '{"file_base64":"…","options":{"includeTables":true}}'

MCP: extract.pdf @ POST https://api.callmeter.dev/mcp · HTTP: POST https://api.callmeter.dev/v1/agentutils/pdf/extract · unpaid → x402 402

Not claimed here

No OCR completeness SLA; scanned image-only PDFs may return thin text. Not a receipt/invoice parser.