You need to choose correctly among text PDFs, scanned PDFs, tables, forms, metadata, or a remote URL.
Its source directly supplies a feature-based decision route and covers all included task areas.
Editorial confidence: highExtract and structure data from PDFs
Compare PDF workflows by text, table, form, OCR, batch, privacy, and structured-output requirements.
Last meaningful review July 20, 2026
The job
You have one or more PDFs and need their text, tables, fields, or metadata extracted into a reliable structure for review or downstream work.
Includedtext and metadata extraction · table or form extraction · batch PDF processing and OCR-aware workflows
Not this pagelegal or medical interpretation · PDF visual design · accuracy claims without source verification
30-second route
Consider Ocr And Documents ↗
You need CSV/Excel tables, JSON form fields, validation, or robust automation controls.
It documents dedicated scripts for these outputs and explicit operational safeguards.
Editorial confidence: highYour primary output is clean Markdown for review or downstream LLM/text processing, including streaming or batches.
Its source directly documents PDF-to-Markdown conversion, stream input, OCR-aware setup, and batch examples.
Editorial confidence: highThe request asks for legal or medical interpretation rather than extraction into a reviewable structure.
Interpretation is outside the stated task scope.
Editorial confidence: highBefore you install
Install one of the selected Skills; do not build a custom workflow first.
All three directly perform PDF data extraction with usable, explicit source-backed instructions. Select based on required output and operational constraints.
Read the intervention guide →Source-led comparison
What changes the choice
Facts are extracted from each selected Skill’s declared workflow. Signals describe public adoption; they are not a universal score.
SkillBest forActionEdits?Setup
Ocr And DocumentsDefault choice across local or remote PDFs, including text, tables, forms, metadata, OCR, and batch work.Install and follow the extractor decision path: web_extract for URLs; pymupdf for ordinary local PDFs; marker-pdf for OCR, forms, equations, or complex layouts.None.Requires python3; the skill declares pymupdf4llm installation. marker-pdf is an on-demand additional install for advanced OCR/layout work. ↗Pdf Processing ProProduction-style extraction that needs CSV/Excel tables, form-field JSON, validation, logging, error handling, and batch automation.Install the listed Python dependencies, then use the included table, text, form-analysis, and validation scripts.None.Install pdfplumber, pypdf, pillow, pytesseract, and pandas; OCR additionally requires the tesseract system package. ↗MarkitdownConverting PDFs to clean Markdown for review or LLM/downstream text workflows, with streaming and directory batch conversion.Install MarkItDown with PDF support (or all features), convert PDFs through the CLI or Python API, and use stream conversion or the documented directory loop for larger/batch input.None.Install `markitdown[pdf]` or `markitdown[all]`; scanned-document OCR requires Tesseract. OpenRouter configuration is optional for LLM-powered features. ↗Source notes
Inspect evidence and candidate limits 3 candidates
Why these candidatesEach candidate comes from tracked public sources and has a source-attributed adoption signal. This is a shortlist, not a catalogue or global ranking.
What this cannot establishWe read declared instructions and public signals. We do not run skills, benchmark outputs, audit security, or claim a universal best choice.
Use it forDefault choice across local or remote PDFs, including text, tables, forms, metadata, OCR, and batch work.
Avoid ifYou cannot accommodate marker-pdf’s stated multi-GB install when OCR or complex-layout extraction is needed.
ActionInstall and follow the extractor decision path: web_extract for URLs; pymupdf for ordinary local PDFs; marker-pdf for OCR, forms, equations, or complex layouts.
EditsNone.
SetupRequires python3; the skill declares pymupdf4llm installation. marker-pdf is an on-demand additional install for advanced OCR/layout work.
Inspect workflow and compatibility details
ContextRoutes by PDF type and required output rather than assuming one extractor fits every document.
OutputPlain text, Markdown, tables, extracted images, metadata, or marker-pdf JSON with metadata.
ControlDecision tree plus helper scripts and CLI commands.
Invocationpython scripts/extract_pymupdf.py document.pdf --tables; python scripts/extract_marker.py document.pdf --json; marker /path/to/folder --workers 4
Install scopeskill dependencies
Agent fitHigh: explicit routing, concrete commands, and stated trade-offs make it suitable for reliable agent-led extraction.
Watch outmarker-pdf is documented as roughly 3–5 GB to install and also downloads models on first use.
Source-visible capabilitiesPDF routingUses web extraction for URLs, pymupdf for local text PDFs, and marker-pdf for scanned PDFs.explicit sourceStructured extractionExtracts tables, metadata, forms, and JSON output.explicit sourceOCR and batchMarker-pdf handles scanned documents and a folder batch command is documented.explicit source
CompatibilityPython environmentsdeclarednot runtime-testedLocal disk-constrained environmentslimitednot runtime-tested
Evidence
The skill covers PDFs and scanned documents and directs URL, local-text, and OCR extraction routes.SKILL.md:16-35 · inspected 2026-07-20 · high confidence
Its capability comparison explicitly includes tables, forms, OCR, Markdown output, and installation/speed trade-offs.SKILL.md:37-57 · inspected 2026-07-20 · high confidence
Documented commands extract tables, images, metadata, page ranges, JSON, and batch folders.SKILL.md:61-75 · inspected 2026-07-20 · high confidence
Marker-pdf commands cover OCR and folder batch processing.SKILL.md:89-111 · inspected 2026-07-20 · high confidence
Public signal282 /SkillsMP source-specificbest-defaulttexttablesformsocrmetadatabatchexplicit source
Use it forProduction-style extraction that needs CSV/Excel tables, form-field JSON, validation, logging, error handling, and batch automation.
Avoid ifYou only need quick document-to-Markdown conversion or want to avoid installing its Python/OCR dependencies.
ActionInstall the listed Python dependencies, then use the included table, text, form-analysis, and validation scripts.
EditsNone.
SetupInstall pdfplumber, pypdf, pillow, pytesseract, and pandas; OCR additionally requires the tesseract system package.
Inspect workflow and compatibility details
ContextA script-oriented workflow focused on structured exports and operational controls.
OutputText files, CSV/Excel tables, JSON form fields, and validated PDF outputs.
ControlPurpose-built CLI scripts with validation, logging, exit codes, and batch examples.
Invocationpython scripts/extract_tables.py input.pdf --output tables.csv --format csv|excel; python scripts/analyze_form.py input.pdf --output fields.json
Install scopeskill dependencies
Agent fitHigh for repeatable downstream pipelines because failure handling and automation behavior are explicitly documented.
Watch outOCR requires a separate tesseract system installation; source does not make a performance guarantee.
Source-visible capabilitiesTable exportsExtracts tables with automatic column detection to CSV or Excel.explicit sourceFormsAnalyzes form fields to JSON and validates filling data.explicit sourceOperational controlsProvides error handling, validation, logging, exit codes, batch operations, and large-file streaming guidance.explicit source
CompatibilityPython environmentsdeclarednot runtime-testedmacOS, Ubuntu, Windows OCR hostsdeclarednot runtime-tested
Evidence
The skill is positioned for forms, tables, OCR, validation, and batch operations.SKILL.md:1-8 · inspected 2026-07-20 · high confidence
It documents table-to-CSV extraction and form-field JSON extraction.SKILL.md:22-41 · inspected 2026-07-20 · high confidence
It explicitly documents error handling, validation, logging, CLI help, exit codes, and batch operations.SKILL.md:43-61 · inspected 2026-07-20 · high confidence
It documents CSV/Excel tables, OCR capabilities, dependency installation, and streaming for large PDFs.SKILL.md:76-96 · inspected 2026-07-20 · high confidence
Dependencies and the extra tesseract requirement are specified.SKILL.md:222-245 · inspected 2026-07-20 · high confidence
Public signal879 /SkillsMP source-specificproduction-workflowcsv-exceljson-formsvalidationocrbatchexplicit source
Use it forConverting PDFs to clean Markdown for review or LLM/downstream text workflows, with streaming and directory batch conversion.
Avoid ifYou require dedicated CSV/Excel table export or a specialized PDF-form workflow.
ActionInstall MarkItDown with PDF support (or all features), convert PDFs through the CLI or Python API, and use stream conversion or the documented directory loop for larger/batch input.
EditsNone.
SetupInstall `markitdown[pdf]` or `markitdown[all]`; scanned-document OCR requires Tesseract. OpenRouter configuration is optional for LLM-powered features.
Inspect workflow and compatibility details
ContextA high-public-recognition alternative focused on Markdown conversion across document types rather than dedicated PDF field/table pipelines.
OutputMarkdown text; stream conversion is supported through the Python API.
ControlCLI and Python API with optional enhanced PDF conversion and plugins.
Invocationmarkitdown document.pdf -o output.md; md.convert_stream(f, file_extension=".pdf")
Install scopePython package and optional OCR/LLM dependencies
Agent fitHigh for straightforward conversion and batch ingestion; scope is strongest when Markdown is the desired structure.
Watch outIts documented OCR dependency is Tesseract; optional LLM-powered features use an OpenRouter API key.
Source-visible capabilitiesPDF conversionConverts PDFs with full text extraction into structured Markdown.explicit sourceOCR-aware processingDocuments OCR for scanned documents and instructs users to install OCR dependencies.explicit sourceStreaming and batchProvides a PDF stream API and a loop that processes every PDF in a directory.explicit source
CompatibilityPython environmentsdeclarednot runtime-testedmacOS and Ubuntu OCR hostsdeclarednot runtime-testedOptional LLM-enhanced featuresdeclarednot runtime-tested
Evidence
MarkItDown supports PDF conversion, OCR for images/scanned documents, and structured Markdown output.SKILL.md:10-22 · inspected 2026-07-20 · high confidence
The supported-format table lists PDF full-text extraction and image OCR.SKILL.md:59-75 · inspected 2026-07-20 · high confidence
The source provides pip installation, CLI PDF conversion, and a Python PDF stream API.SKILL.md:77-122 · inspected 2026-07-20 · high confidence
It documents processing all PDFs in a directory and batch conversion with error handling.SKILL.md:230-299 · inspected 2026-07-20 · high confidence
It recommends OCR dependencies for scanned documents and streaming for large PDFs.SKILL.md:323-362 · inspected 2026-07-20 · high confidence
Public signal49 /SkillsMP source-specificmarkdown-conversionhigh-recognitionocr-awarestreamingbatchexplicit source
Still comparing?
Tell us the job, not just the skill names.
Request a task guide ↗Did this decision map help?