How do I install the PDF Backends skill?

Run npx skillmds add xberg-io/pdf-backends in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is PDF Backends safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with PDF Backends?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is PDF Backends free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published PDF Backends?

xberg-io (@xberg-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

PDF Backends by xberg-io — Agent Skill

Name: PDF Backends
Rating: 4.666666666666667 (3 reviews)
Author: xberg-io

PDF Backends

Change or diagnose Xberg PDF extraction, native/Pdfium backend selection, PDF rendering sessions, encrypted documents, OCR fallback, or backend-specific capability gaps. Load for PDF engine work, not generic image OCR.

xberg-io b2ede75 2.2 KB Updated Sep 9, 2026

File contents

PDF backends

PdfConfig.backend selects PdfBackend::Native (default) or PdfBackend::Pdfium. Backend choice is explicit: never silently return native output when Pdfium was requested.

Native backend

The native engine is xberg_native_pdf and supports the full Xberg PDF pipeline: text and structure extraction, page rendering/OCR fallback, tables, annotations, images, attachments, and richer metadata.
config.force_ocr forces the whole document; config.force_ocr_pages selects pages. Otherwise, pages without usable native text can enter the OCR stage inside PdfExtractor; OCR is not a second PDF extractor in the registry.
PdfRenderSession in pdf/render.rs opens document bytes once and exposes page count and page rendering. Use it for repeated page renders instead of reopening the same document for each page. It is a Rust primitive and need not be exposed by every binding.

Pdfium backend

Pdfium requires the pdf-pdfium Cargo feature. Selecting it without that feature must return an actionable error.
Runtime dynamic loading honors PDFIUM_DYNAMIC_LIB_PATH as a directory containing the platform library. Build-time static and dynamic link variables are not interchangeable with runtime discovery.
Pdfium binding is process-global and operations are serialized by the engine lock. Preserve that invariant when changing concurrency.
The Pdfium implementation intentionally has a narrower capability surface: text, page count, and Info-dictionary metadata. It warns about omitted native capabilities rather than pretending parity.

Verification

Test backend selection separately from document semantics. A missing Pdfium runtime is infrastructure failure unless the job promises to provision it; when XBERG_REQUIRE_PDFIUM is set, skipping because the library or fixture is absent is a failure. Use corpus fixtures through the test-corpus skill and include Pdfium artifacts in benchmark aggregation when the adapter is enabled.

---name: pdf-backendsdescription: Change or diagnose Xberg PDF extraction, native/Pdfium backend selection, PDF rendering sessions, encrypted documents, OCR fallback, or backend-specific capability gaps. Load for PDF engine work, not generic image OCR.--- # PDF backends `PdfConfig.backend` selects `PdfBackend::Native` (default) or `PdfBackend::Pdfium`. Backend choice is explicit: never silently return native output when Pdfium was requested. ## Native backend - The native engine is `xberg_native_pdf` and supports the full Xberg PDF pipeline: text and structure extraction, page rendering/OCR fallback, tables, annotations, images, attachments, and richer metadata. - `config.force_ocr` forces the whole document; `config.force_ocr_pages` selects pages. Otherwise, pages without usable native text can enter the OCR stage inside `PdfExtractor`; OCR is not a second PDF extractor in the registry. - `PdfRenderSession` in `pdf/render.rs` opens document bytes once and exposes page count and page rendering. Use it for repeated page renders instead of reopening the same document for each page. It is a Rust primitive and need not be exposed by every binding. ## Pdfium backend - Pdfium requires the `pdf-pdfium` Cargo feature. Selecting it without that feature must return an actionable error. - Runtime dynamic loading honors `PDFIUM_DYNAMIC_LIB_PATH` as a directory containing the platform library. Build-time static and dynamic link variables are not interchangeable with runtime discovery. - Pdfium binding is process-global and operations are serialized by the engine lock. Preserve that invariant when changing concurrency. - The Pdfium implementation intentionally has a narrower capability surface: text, page count, and Info-dictionary metadata. It warns about omitted native capabilities rather than pretending parity. ## Verification Test backend selection separately from document semantics. A missing Pdfium runtime is infrastructure failure unless the job promises to provision it; when `XBERG_REQUIRE_PDFIUM` is set, skipping because the library or fixture is absent is a failure. Use corpus fixtures through the `test-corpus` skill and include Pdfium artifacts in benchmark aggregation when the adapter is enabled.

Frequently asked questions