classify_pdfFast
Classify a PDF as text-based, scanned, image-based, or mixed, with confidence and OCR-page guidance.
url or base64Classify before OCR
A fast first pass keeps document pipelines efficient and sends only scanned pages to OCR or vision models.
01 · CLASSIFY
Determine whether the document is text-based, scanned, image-based, or mixed.
02 · ROUTE
Use the returned OCR page list to avoid processing native-text pages unnecessarily.
03 · EXTRACT
Choose plain text, positioned text, or layout-aware Markdown for downstream work.
Tools
Every tool accepts either a public PDF URL or base64-encoded PDF bytes.
detect_pdfMetadata
Return document type, page count, and pages needing OCR without producing Markdown.
url or base64process_pdfFull parse
Parse the document into text and Markdown with type and layout metadata.
url or base64 · optional pages[]pdf_to_markdownMarkdown
Produce clean per-page Markdown plus table, column, and OCR layout flags.
url or base64 · optional pages[]extract_textText
Extract plain text for search, summarization, indexing, and analysis.
url or base64extract_text_with_positionsLayout
Return positioned text items with font and bounding-box metadata.
url or base64 · optional pages[]Streamable HTTP
Use the custom-domain MCP endpoint with any non-empty Bearer token.
{
"mcpServers": {
"pdf-inspector": {
"type": "streamable-http",
"url": "https://pdfinspectormcp.agenticledger.ai/mcp",
"headers": { "Authorization": "Bearer pdf" }
}
}
}