One Engine, Seven Open-Source Projects: PageSift 1.4 and the yolow Family A developer released PageSift 1.4, completing a refactor that splits the scraper's engine into seven independently versioned MIT-licensed open-source projects, including yolowdoc, yolowsyntax, yolowzstd, yolow64, yolowpdf and yolowhtml. The release adds DataGraph, a ~4,000-line crate that tracks provenance, conflicts and versioning of extracted web data for LLM pipelines, exposed via CLI, HTTP API, an MCP tool and Python/Node SDKs. The developer reports benchmarks including 20.4 GB/s Base64 throughput, 1,900 pages/s PDF-to-Markdown conversion, and 1.000 structural F1 for Markdown extraction. PageSift started as the ingestion layer of an AI research platform I was building: something had to fetch web pages and documents and turn them into clean Markdown that an LLM pipeline could actually use. It grew into a self-hosted web scraper, crawler and document parser — Firecrawl- and Jina-compatible API, MCP server, Python/Node SDKs, one Rust binary. Over the 1.x series I kept doing one thing: every time a piece of the engine turned out to be generally useful, I cut it out and gave it its own repository, its own versioning, and an MIT license. PageSift 1.4 finishes that split — and adds the subsystem the extraction was making room for. The split, completed. Wave one 1.2.7 turned the base64, PDF, HTML and content crates into thin shims over standalone libraries. Wave two 1.4.0 does the same for the remaining three: pagesift-office → yolowdoc , pagesift-syntax → yolowsyntax , pagesift-compress → yolowzstd . The engine is now a composition of seven independently versioned open-source projects — and since 1.3, PageSift itself is Apache-2.0 it was AGPL before . DataGraph. The new crate, pagesift-graph ~4,000 lines , is what the cleanup was for. A normal scraper returns text and forgets everything else. DataGraph keeps the receipts: sku are clustered into It's exposed everywhere PageSift is: CLI scrape --graph , crawl --graph , graph inspect , the HTTP API a graph slice on scrape responses, plus /v2/graph/history and /v2/graph/conflicts , the MCP tool pagesift graph query , and the Python and Node SDKs. An LLM pipeline doesn't just need text. It needs to know where a claim came from, what it contradicts, and which version is current. That's the layer DataGraph adds. A fast Base64 codec with runtime-dispatched SIMD AVX2/NEON , zero runtime dependencies, streaming and data-URL APIs. The 0.3 release added dedicated short-input kernels and fused hash-table tail processing: around 20.4 GB/s from 8 KiB to 1 MiB , and short payloads 8–64 bytes got 1.5–2.8× faster than in 0.2. Short inputs are the common case in real pipelines — tokens, hashes, embedded blobs — so that's where the work went. cargo add yolow64 · GitHub https://github.com/YOLOW92/yolow64 · crates.io https://crates.io/crates/yolow64 A PDF-to-Markdown library and OCR engine that parses the format itself — no Poppler, no MuPDF, nothing to link against. The workspace includes GPU-accelerated OCR PP-OCRv6 . The README benchmark clocks 1,900 pages/s on born-digital documents i7-10700 . This is the component that taught the family its hardest rule: every allocation a hostile file can influence is budgeted before it happens. cargo add yolowpdf · GitHub https://github.com/YOLOW92/yolowpdf · crates.io https://crates.io/crates/yolowpdf Two crates: yolowhtml-dom , a zero-copy HTML5 parser over a contiguous arena Vec