Week 45, updated 2 Nov 2026

AI use cases Research & web data Convert documents to Markdown for LLMs

Convert documents to Markdown for LLMs

Turn PDFs, Word, Excel and PowerPoint files into Markdown that AI models can read and search. Prompts, workflow and ranked converters.

Tools compared
4
Open source
4of 4
Reviewed
16 Oct

For: Developers · Researchers and analysts · Small business

Best-scored tools for this jobscore, safety
  1. 1FirecrawlAI tool78No issues found
  2. 2MarkItDownAI tool54No dependencies scanned
  3. 3Crawl4AIAI tool52Caution
  4. 4LlamaIndexAgent framework43Caution
What the score and shapes mean

Score = public maintenance, adoption and docs signals, as a percentile of 52 tools. Nobody here tested these hands-on. Shapes are the separate safety label: No issues found, No dependencies scanned, Caution.

What it does for you

Anyone preparing internal documents for AI search, summarising or analysis.

  • Convert Word, Excel, PowerPoint and PDF files to Markdown
  • Batch-convert a folder of documents
  • Prepare files for chunking and search
  • Keep table structure for analysis by a model
  • Feed documents to an agent through a command line or script
More about this job

Most business knowledge sits in office files. Converting them to Markdown keeps headings, lists and tables in a format models handle well, and it is the first step of most document-based AI projects.

Workflow in 5 steps

  1. Collect a representative sample of documents, including the ugly ones.
  2. Convert and compare a few against the originals.
  3. Note what is lost and decide if OCR is needed.
  4. Batch-convert and keep the original path as metadata.
  5. Add a re-run step when files change.
Full steps
  1. Collect a representative sample of documents, including the ugly ones.
  2. Convert and compare a few against the originals.
  3. Note what is lost (images, complex layouts, scanned pages) and decide if OCR is needed.
  4. Batch-convert and keep the original path as metadata.
  5. Add a re-run step when files change.

Prompts to copy

Starting points we wrote, not tested results.

Write a Python script that converts every .docx and .pdf in ./docs to Markdown in ./md and prints the files that failed.

2 more prompts

Check this converted Markdown against the original for lost tables, headings or footnotes and list what is missing.

Split this Markdown into sections by heading and give each section a one-line summary.

Watch out for

  • Prompts sent to a hosted model leave your machine.
  • AI output can be fluent and wrong.
  • Scanned PDFs often need OCR, which converters may not do.
  • Licence to check before commercial use: Firecrawl (AGPL-3.0).
  • No issues foundFirecrawlNo issues found in this week’s scan
  • No dependencies scannedMarkItDownCode checks clean, no lockfile to check
  • CautionCrawl4AI29 critical-severity known vulnerabilities in runtime dependencies (OSV)
  • CautionLlamaIndex39 critical-severity known vulnerabilities in runtime dependencies (OSV)
Why these warnings?
  • Prompts sent to a hosted model leave your machine. For confidential material use a local model or check the provider's data terms first.
  • AI output can be fluent and wrong. Check facts, code and citations yourself before you rely on them or publish them.
  • Scanned PDFs often need OCR, which converters may not do. Check output on scanned samples before trusting a pipeline.
  • MarkItDown: code and instruction-file checks found nothing, but there is no runtime lockfile or manifest, so dependencies were not checked
  • Crawl4AI: 29 critical-severity known vulnerabilities in runtime dependencies (OSV)
  • LlamaIndex: 39 critical-severity known vulnerabilities in runtime dependencies (OSV)

Recommended tools

From our ranked data, highest score first. A high score does not mean a clean scan.

Related jobs

Sources for tool facts

firecrawl/firecrawl · microsoft/markitdown · unclecode/crawl4ai · run-llama/llama_index · MarkItDown on GitHub. Data fetched 2 Nov 2026. Scores and labels are automated, relative and not a hands-on test.

Get the board every Monday

One email a week: the new top 10, the biggest score changes and every safety label that changed.

Free. Double opt-in: you confirm by email first, unsubscribe in one click.