Markitdown

Skill matematicsolutions/awesome-matematic-skills-pl/dokumenty/skills/markitdown

Polish hub of legal-AI skills - 41 skills in 8 Claude Code / Cowork bundles (LLM output verification, PL/EU case law, DOCX redline, AI Act audit bundle). GDPR-safe, vendor-neutral, MIT.

Install

npx -y skills add matematicsolutions/awesome-matematic-skills-pl --skill markitdown

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Konwersja dowolnego dokumentu (PDF, Word, Excel, PowerPoint, HTML, EPUB, audio, obrazy, YouTube) na Markdown dla LLM. Użyj gdy użytkownik mówi "konwertuj PDF", "przerób Word na markdown", "zamień PPT na MD", "markdown z Excela", "wyciągnij tekst z PDF", albo daje plik Office/PDF do analizy. Microsoft MarkItDown (pip) + MCP server.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

2.6 KB, 731 tokens by cl100k_base, as published. Nobody here has run it

MarkItDown - konwerter dokumentów do Markdown (PL)

Lightweight utility Microsoftu - zachowuje strukturę (nagłówki, listy, tabele, linki), nie wygląd. Pod LLM, nie pod human.

Instalacja (zrobione 2026-04-21)

python -m pip install --user markitdown markitdown-mcp

Wymaga Python 3.10+ (testowane na 3.14). CLI: python -m markitdown.

Wspierane formaty

PDF (preferuj dla krótkich, standardowych PDF; dla złożonych/tabel - OpenDataLoader PDF)
Office: Word (.docx), Excel (.xlsx), PowerPoint (.pptx)
HTML, EPUB, CSV, JSON, XML
Obrazy (EXIF + OCR jeśli zainstalowane [all])
Audio (EXIF + transkrypcja jeśli włączone)
ZIP (iteruje zawartość)
YouTube URL (napisy)

Użycie

CLI (single file)

python -m markitdown input.pdf > output.md
python -m markitdown input.pptx -o output.md

Batch (Obsidian Vault)

for f in "/c/Users/hp/Documents/Obsidian Vault/Konwerter"/*.pdf; do
  python -m markitdown "$f" > "${f%.pdf}.md"
done

Python API

from markitdown import MarkItDown
md = MarkItDown()
result = md.convert("plik.docx")
print(result.text_content)

MCP server

Opcjonalnie - jeśli chcesz udostępnić Claude Code jako MCP tool:

markitdown-mcp

Kiedy użyć MarkItDown vs OpenDataLoader PDF

Sytuacja	Narzędzie
Word/Excel/PPT	MarkItDown
Prosty PDF, tekst liniowy	MarkItDown (szybsze)
Złożony PDF z tabelami, reading order, papers naukowe	OpenDataLoader PDF (jakość)
Audio/transkrypcja	Whisper (mamy `whisper-asr-pipeline`)
Web page	Defuddle (mamy `defuddle` od Kepano)

Integracja z vault

Output docelowo → Konwerter/ (folder w vault dla source-pdf → MD). Frontmatter: type: source-pdf, tags: [pdf, zrodlo]. Zobacz vault-rules.json → clippings.classification_rules.pdf_source.

Gotcha

markitdown[all] na Python 3.14/3.15 failuje (onnxruntime, youtube-transcript-api konflikt) - użyj zainstalowanego markitdown bez extras
Dla OCR obrazów / pełnego YouTube - Docker albo starszy Python (3.11/3.12)
Nie używaj dla high-fidelity conversion dla human reading - tylko pod LLM

Gives 0 of the 12 instructions most mcp tooling skills give in 731 tokens

Counted across 638 of the 750 authors here whose files we hold, read 2026-08-06

create ten complex read-only evaluation questionsin 71 of 638, across 17 files
test servers using MCP Inspectorin 60 of 638, across 18 files
provide actionable error messagesin 56 of 638, across 14 files
prioritize comprehensive API coverage over specific workflowsin 54 of 638, across 12 files
use TypeScript and Streamable HTTP for remote serversin 53 of 638, across 7 files
define structured output schemas where possiblein 51 of 638, across 9 files
use Zod or Pydantic for input schemasin 48 of 638, across 6 files
fetch MCP specification pages with markdown suffixin 46 of 638, across 4 files
load framework documentation using WebFetchin 45 of 638, across 3 files
verify each evaluation answer independentlyin 45 of 638, across 3 files
implement API client with authentication and paginationin 45 of 638, across 3 files
Define input schemas with validationin 28 of 638, across 10 files

Said here and by no other author read

use markitdown without extras
use OpenDataLoader PDF for complex PDFs
use Whisper for audio transcription
use Defuddle for web pages
store converted Markdown output in the vault
add frontmatter to converted Markdown files

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.