pdf-inspector

pdf-inspector is a Rust-powered library that classifies PDFs and extracts native text into clean, position-aware Markdown. It detects whether PDFs are text-based, scanned, image-based, or mixed, then routes them appropriately—skipping expensive OCR for documents that don't need it. The tool handles multi-column layouts, tables, font decoding, and semantic formatting while keeping processing under 200ms for typical documents.

Available as npm, PyPI, and crates.io packages with a bundled CLI, plus browser WebAssembly support. The project describes itself as built by Firecrawl to handle text-based PDFs locally without relying on external OCR services. Developers can use it from Node.js, Python, Rust, or directly in browsers.