What specific types of document formatting can this skill preserve during conversion?

This skill converts PDF, Word, Excel, and CSV files into structured GitHub Flavored Markdown (GFM), preserving complex elements like nested lists, merged table cells, and reading order while using OCR for scanned documents.

How does the skill handle unreadable text or low-quality scans?

Unlike standard converters, this skill prioritizes accuracy over completion; it flags low-confidence OCR regions and explicitly notes unreadable charts or missing pages in a YAML frontmatter instead of 'hallucinating' or inventing text.

How do I use File to Markdown: PDF, Word, Excel & CSV Converter?

Connect your AI agent to Agensi through MCP, then ask it to use File to Markdown: PDF, Word, Excel & CSV Converter on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

Can this skill handle multi-sheet spreadsheets and merged cells in tables?

Yes, the skill handles multi-sheet Excel workbooks by converting them into sequential Markdown sections and manages complex table structures like rowspans and colspans by falling back to HTML table tags when necessary to maintain data integrity.

What is included with my purchase of the File to Markdown skill?

Upon purchase, you receive the full SKILL.md definition file, comprehensive documentation for integration, and lifetime access to any future version updates or performance improvements.

Does the skill support different text encodings for CSV and TXT files?

The converter includes a sophisticated encoding fallback system that automatically tries UTF-8, BOM, cp1252, and latin-1 to ensure CSV and text files are read correctly without manual intervention.

File to Markdown: PDF, Word, Excel & CSV Converter

Example session with this skill installed

Convert the attached Q3_Revenue_Report.pdf to markdown. It has three pages and one table on the second page. Just give me the markdown content.

Read your context and instructions
Compiled the file to markdown

I have processed Q3_Revenue_Report.pdf. The file contains a text layer across all 3 pages. One pipe table was detected and extracted from page 2.

Q3_Revenue_Report.md


# Q3 2024 Revenue Report

## Executive Summary
Quarterly performance exceeded targets by 8% despite seasonal headwinds in the retail sector. Growth was largely driven by the expansion of the enterprise software division.

## Regional Performance
The following table outlines the revenue breakdown by territory.

| Territory | Target ($M) | Actual ($M) | Variance (%) |
|:---|---:|---:|---:|
| North America | 45.0 | 48.5 | +7.7% |
| EMEA | 32.0 | 35.2 | +10.0% |
| APAC | 28.5 | 29.1 | +2.1% |
| LATAM | 12.0 | 11.8 | -1.6% |


## Strategic Initiatives
1. **Cloud Migration**: Completed for 90% of legacy clients.
2. **Channel Partners**: Added 15 new resellers in the EMEA region.
3. **Product Launch**: The 'Vanguard' suite is scheduled for Q1 2025.

[^1]: Variance calculated against adjusted Q2 baseline.

Next steps

I can re-run this with HTML table tags if you need to preserve specific cell padding or alignment.
Would you like me to extract the summary into a separate JSON file for your database?

I have processed Q3_Revenue_Report.pdf

Connects securely to your tools. The creator never sees your data.

About this skill

Turn any document into Markdown you can actually use — in your wiki, README, RAG pipeline, or note-taking workflow.

What it converts: PDF (text-layer and scanned), Word (.docx/.doc), Excel (.xlsx/.xls), CSV/TSV, TXT/MD/RTF.

What you get: Clean GFM Markdown with preserved heading hierarchy, pipe tables (or HTML tables for merged cells), nested lists, code blocks, hyperlinks, and Markdown footnotes. Every output file is self-describing via YAML frontmatter — source filename, page count, table count, and a notes: field documenting any inference or correction.

Why it's different

OCR with honesty. Scanned PDFs are processed at 300 DPI with per-page confidence reported. Regions under 70% confidence are flagged — never silently transcribed. If OCR fails entirely, you get an image description and a warning, not fabricated text.

No hallucination. If a page is missing, a table is unreadable, or a chart can't be parsed, it says so. Every output traces back to the source.

No silent drops. Every page, sheet, and section is either converted or explicitly noted as skipped.

Deterministic. Same input + same options → same Markdown. Reproducible across runs.

Edge cases handled: Multi-sheet workbooks, merged cells (rowspan/colspan), lettered lists, page-count mismatches, running headers/footers, DOCX tracked changes, CSV encoding fallbacks (UTF-8 → BOM → cp1252 → latin-1), multi-column reading order, and password-protected files (asks for the password rather than cracking).

Ideal for: developers ingesting docs into RAG, writers migrating notes, legal/ops teams converting archives, anyone who needs a document as text without losing structure.

Install: Drop the SKILL.md into ~/.claude/skills/file-to-markdown/. Claude Code picks it up automatically. Invoke with /file-to-markdown or just upload a file and say "convert this to Markdown."

How to use

Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

Reviews

No reviews yet

Be one of the first to try it. Every listed skill passes our trust checks below.

Security scanned

Passed our 8-point scan before listing

Fresh listing

Recently published to Agensi

30-day refund

Not a fit? Get your money back

File to Markdown: PDF, Word, Excel & CSV Converter